Fri Aug 20 15:35:23 1993 =============================================================================== Subject: pvm survey To: pvm-survey-list = those who have requested PVM 3 src from netlib@ornl. Dear Colleague, In order to improve PVM we are trying to understand how people are using the system. Could you take a moment to answer three questions. Are you using PVM today? What are your application(s)? How many and what kind of machines have you used with PVM? Any comments and suggestions for improvement are welcome. Please send the information to pvm-survey@cs.utk.edu. Your help is much appreciated. Best wishes, The PVM Project Team =============================================================================== From: Gary Hanyzewski >| >| Are you using PVM today? Personally.. No. Institutionally.. Yes. >| >| What are your application(s)? Distributed Global Change Modeling >| >| How many and what kind of machines have you used with PVM? Networks of Sun Sparcstations, SGI's ( everything from indigos and personal Iris's to Challenge Series multi-processor reality engine machines ), NEXT, DECstation ( 5000 and 3000 ), Cray Y-MP, Intel iPSC/860 and Intel Paragon. >| >| Any comments and suggestions for improvement are welcome. >| Please send the information to pvm-survey@cs.utk.edu. >| Your help is much appreciated. Additional support for an better integration of MPP PVM versions with non MPP code. I am still having some small trouble in getting PVM 3.1 to work correctly on our PARAGON. >| >| Best wishes, >| The PVM Project Team And to you. PVM is a great tool. Gary H. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Gary Hanyzewski garyh@sdsc.edu San Diego Supercomputer Center (619) 534-5129 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ------------------------------------------------------------------------------- From: heistand@iastate.edu Dear Sirs, I am quite happy to fill out your survey as I wish PVM to become as powerfull and widespread as possible. I am using PVM to obtain a numerical solution to a 3-d fluids problem, basically the flow around two aircraft wings in close proximity. I am very impressed and pleased with PVM. I am running the model on DEC's, approx. 50 machines, most of them being 5000/25 and a few 5000/133 and 5000/240's. Steve Heistand ------------------------------------------------------------------------------- From: Roberto.Bisiani@N3.SP.CS.CMU.EDU Are you using PVM today? YES What are your application(s)? Speech recognition How many and what kind of machines have you used with PVM? DEC 3000, up to 5, Ethernet (the application really needs a much faster network that I will get in the future). Any comments and suggestions for improvement are welcome. My application exchanges lots of data. I am often getting into trouble with lack of memory. I am currently frustrated enough that I have started looking at pvm sources to understand what is going on. I only use the direct communication option since communicating through the demon does not make much sense in my case and only worsens performace and memory allocation problems. Roberto Bisiani ------------------------------------------------------------------------------- From: nrl@SSESCO.com > > Are you using PVM today? Yes, as we speak. > What are your application(s)? Meteorological, Photochemical,and Hrdologicall models which heretofor could only be run on LARGE Cray machines > > How many and what kind of machines have you used with PVM? At this site we regularly run on from 2 to 7 RS6000s. At times we use 2 SGI and two SPARC systems. At FSU we have used up to sixteen machines for runs of up to six hours. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. Still waiting to hear from Al about my temporary implementation of the "speed" parameter in the host data structure... ------------------------------------------------------------------------------- From: wcs@nas.nasa.gov (William C. Saphir) >Could you take a moment to answer three questions. (I am one of the main contacts at NAS for people using (or thinking of using) PVM. The information below includes myself and the people with whom I have regular contact. Does not include people who are just starting (several people are thinking about using PVM for CFD applications) I may have missed a few. Hopefully they are on your list). ----------------------- > >Are you using PVM today? Yes. > >What are your application(s)? >How many and what kind of machines have you used with PVM? 1. NAS parallel benchmarks (tightly coupled. lots of communication) PVM 2.4. Has been run on SGI workstations. Poor speedup. Being ported to 3.1, mpps (?) Contact: Raad Fatoohi (fatoohi@nas.nasa.gov) 2. Medusa (CFD application. uses domain decomposition with overset grids so that communication requirements aren't too large. manager-worker model. Run on SGI workstations. Good speedup. Contact: Merritt Smith. (mhsmith@nas.nasa.gov) 3. Another cfd application. Also manager-worker. Run on Intel Hypercube (and network of workstations?). Good speedup. Contact: Graham Rhodes (sues@nas.nasa.gov) 4. Simple benchmarks to measure PVM/mpp performance vs. native mpp message passing. Contact: me On number of workstations: Order of 10. iPSC/860 nodes: up to 128. > >Any comments and suggestions for improvement are welcome. >Your help is much appreciated. In no particular order and of varied importance. 1. For many sites, instructions (and a simple procedure) on how to install PVM in a central place (not a home directory) would probably be useful. 2. As I've already commented to a few people, support for global group operations would be useful for us. Dynamic groups do not seem to be too useful for numerical codes. This is something I'm working on. I'll keep you informed of progress. 3. I/O could be a real problem for a lot of cfd applications. I realize it is probably impossible for PVM to do anything about this. People who need to do a lot of I/O will either avoid PVM or write machine-specific code to handle it. 4. Is the "f" in fortran calls necessary? It's ugly. PVM_send() and pvm_send() perhaps? (like MPI). 5. PvmDataInPlace would be really good (although I suppose since multiple packs probably result in multiple sends it should come with a warning). Also, xdr encoding only when necessary would be good. Several people have mentioned these. 6. The FAQ that's floating around is out of date and not very good. 7. Any plans for an instrumented library (or a flag?) to produce traces? 8. A utility library would be useful. Timers are the main thing. I'm planning to produce one, and will keep you informed. >Please send the information to pvm-survey@cs.utk.edu. 9. The survey was sent from "owner-pvm-survey." Although replies go the right place, you may lose some: telnet cs.utk.edu 25 Connected to cs.utk.edu. 220 CS.UTK.EDU Sendmail 5.61+IDA+UTK-930125/2.8s-UTK ready at Tue, 17 Aug 93 22:25:41 -0400 vrfy owner-pvm-survey 250 ---------------------------- You're probably aware of this, but we seem to have 2 distinct categories of PVM users. 1. People making use of idle workstation cycles. (Cheap powerful computing). 2. People who want their codes to be portable. (mostly MPP users). There's a gray area, of course, but it often seems like one reason is dominant. Bill Saphir wcs@nas.nasa.gov ps. I'd be interested in seeing the results of this survey. ------------------------------------------------------------------------------- From: kortas@ctrss1.lanl.gov (Samuel Kortas) yes, we are using PVM. We have a global ocean code here in Los Alamos, developped on the CM-5, and our job is to make it run on other MIMD configurations. Among those, we made it run on an IBM Cluster (1 to 16 nodes), under PVM 2.4, and on a T3D emulator under PVM 3. All our codes are written in fortran (either 77 or CM-Fortran) > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. Yes, PVM is great because we hope that in a long run we'll only need one version of our code for the IBM cluster, but also for the CM-5, the IPSC or the Paragon. But... we had a small problem with the length of a message containing integers. the instructions fputnint and getnint don't seem to work the same everywhere : we always work in double precision (INTEGER*4) and on certain machines, (the IBM Cluster for example), if we want to send/receive and integer message of size n we need to call fputnint (buffer,n*2, info). on other just the regular fputnint(buffer,n,info) will work.... Have you heard of other persons who might have had the same problem? fputnint really sends INTEGER*4 doesn't it? > Best wishes, > The PVM Project Team Thank You for providing a portable environement! |\/\/\/| --------------------------------------------------- | | Samuel Kortas kortas@lanl.gov | | M.S K723 | (o)(o) LOS ALAMOS NATIONAL LABORATORY C _) LOS ALAMOS NM 87544 | ,___| UNITED STATES | / /____\ GRA / T-3 Phone : (505) 665-7782 / \ --------------------------------------------------- ------------------------------------------------------------------------------- From: chuckr@bigbird.llnl.gov (Chuck Rendleman) > > What are your application(s)? Computational fluid dynamics, adaptive mesh refinement. > > How many and what kind of machines have you used with PVM? Massive Parallel systems (CM5, Meiko CS2, Cray T3D, Paragon). Development work on SGI crimson. ------------------------------------------------------------------------------- From: Bill Pearson Are you using PVM today? Extensively What are your application(s)? DNA and protein sequence comparison How many and what kind of machines have you used with PVM? Sun4s: 16, SGI-Indigos: 6, DecALPHA's: 5 ------------------------------------------------------------------------------- From: eric@acis11.tamu.edu Are you using PVM today? Yes we are. What are your application(s)? We are using it to compare network traffic/performance over ethernet, fddi, and token-ring. How many and what kind of machines have you used with PVM? currently, we have 5 IBM rs/6k's (4 320h & 1 550) and 4 sparc-10's (1 4/50 and 3 4/75). Comments: We've just started this work and as yet do not have the fddi/tr up yet. I've heard rumors about trouble with mutilple interfaces and only hope they aren't true???? Hope this helps. Eric ---------- || Eric L. Nelson (Eric.L.Nelson@tamu.edu) _/_/_/_/ || Research Associate _/ _/_/_/ _/ _/_/_/ || Autonomous Underwater _/_/_/ _/ _/ _/ _/ _/ || Vehicle Controller Project _/ _/_/_/ _/ _/ || Texas A&M University _/_/_/_/ _/ _/ _/ _/_/_/ || (409) 845-0085 / (713) 256-1528 ------------------------------------------------------------------------------- From: brown@salt.ra.anl.gov Yes, I am using PVM. Several other poeple in the Reactor Analysis Division are also using it for different applications. (I have no info on PVM usage for the rest of Argonne.) At present, we are still using version 2.4, but expect to switch to 3.1 sometime soon. My applications: Monte Carlo calculations for neutron transport & reactor analysis. Other applications in the Division: Statistical analysis of experimental data; analysis of measured data related to fuel-failure indicators. Planned applications: Neutron diffusion theory. +--------------------------------------------------------+ | Forrest B. Brown || email: fbbrown@anl.gov | | Argonne National Laboratory || voice: (708) 252-9622 | | 9700 S. Cass Avenue - RA/208 || fax: (708) 252-4500 | | Argonne, IL. 60439-4842 || | +--------------------------------------------------------+ ------------------------------------------------------------------------------- From: prange@sdr.slb.com (Michael Prange) > Are you using PVM today? Yes. > What are your application(s)? Numerical modeling of elastic and EM wave propagation. Modeling neutron transport. Solving large matrix systems using blacs. > How many and what kind of machines have you used with PVM? 40 workstations over ethernet and internet. Sparc1+, Sparc2, Sparc10, RS6000, Iris Indigo. I'd like to use our dec alphas running VMS. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Keep up the good work. I'd like to see tools to simplify load balancing. Also, hence is a nice idea, but it needs to include load balancing and more generality. Michael ------------------------------------------------------------------------------- From: rdaniel@acl.lanl.gov (Ronald E. Daniel) > Are you using PVM today? Yes, but I have just started working with it in the last month. > What are your application(s)? My primary interest is in debugging parallel programs. The parallel debugger itself does not use PVM, although I am using Xab3 beta to collect events for tracing execution. For mostly-unrelated work I will be using PVM to implement a constraint satisfaction algorithm based on simulated annealing. > How many and what kind of machines have you used with PVM? Currently, only Sun workstations. In the not-too-distant future I will probably go to a heterogeneous mix of machines - IBM, SGI, and Sun workstations. Ron Daniel Jr. email: rdaniel@acl.lanl.gov Advanced Computing Lab voice: (505) 665-7453 MS B-287 TA-3 Bldg. 2011 fax: (505) 665-4939 Los Alamos National Lab tautology: "Conformity is very popular" Los Alamos, NM, 87545 ------------------------------------------------------------------------------- From: Mark Bolstad Message-Id: <9308161805.AA25464@blad.rtpnc.epa.gov> To: pvm-survey@cs.utk.edu Subject: Re: pvm survey > Are you using PVM today? Yes. > What are your application(s)? Volume Rendering, Data Interpolation, Ray Tracing, Radiosity > How many and what kind of machines have you used with PVM? 11 Machines with 22 Processors (1 each of an 8,4,2 processor machines) SGI Dec Sun Cray Data General I would like to see a port to the MasPar. Mark Bolstad ------------------------------------------------------------------------------- From: htree@gdstech.grumman.com (Hong-Yi Ip) Are you using PVM today? Not today, but may resume in the future. What are your application(s)? image processing, distributed database, etc... How many and what kind of machines have you used with PVM? Suns, DEC ultrix, SGI... Any comments and suggestions for improvement are welcome. Please send the information to pvm-survey@cs.utk.edu. Your help is much appreciated. Best wishes, The PVM Project Team Please send me a summary of this survey if possible. regards --Hong-Yi ***************************************************************************** Hong-Yi Ip (516)346-8724, FAX:(516)575-0965, email: htree@gdstech.grumman.com Grumman Data Systems, addr: MS D12-25, 1111 Stewart Ave, Bethpage, N.Y. 11714 ***************************************************************************** ------------------------------------------------------------------------------- From: guo@meindigo.esrd.com (Guo) >Are you using PVM today? Yes. >What are your application(s)? p-version finite element methods. >How many and what kind of machines have you used with PVM? 4 machines: SGI, RS6K, SUN4, HPPA >Any comments and suggestions for improvement are welcome. NCUBE porting would be useful to us. ------------------------------------------------------------------------------- From: jeg@kaiser.Lanl.GOV (James E Gubernatis) > > Are you using PVM today? Yes. > What are your application(s)? Qunatum Monte Carlo simulations exact diagonalization studies of systems of strongly interacting electrons. > How many and what kind of machines have you used with PVM? Separately on a network 6 Sun workstations and a network of 16 IBM workstations. ------------------------------------------------------------------------------- From: chih@MasPar.COM (Sam Chih) > Are you using PVM today? > Since we received the PVM, we haven't completed the port to our MPP machine yet. > What are your application(s)? > Not decided at this time. > How many and what kind of machines have you used with PVM? > We have not used the PVM yet. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Can you periodically publish the current development or user experience with the PVM ? Thanks. ------------------------------------------------------------------------------- From: Jamie Painter Are you using PVM today? Yes > What are your application(s)? Volume Rendering of 3D Scalar data sets. The volume rendering code originated on a CM-5 using CMMD for message passing and was ported to PVM to allow use in a workstation network environment. > How many and what kind of machines have you used with PVM? We've run up to 32 nodes with the following architectures: HPPA, RS6K, SGI, SUN4. --Jamie ------------------------------------------------------------------------------- From: "Brian Grant" Reply to: RE>pvm survey Yes, I am using PVM3.0 for "coarse-grain decomposition" in solid-state physics codes. For example, if I need to solve for wave functions at numerous k-points, I will use pvm to spread them out. Because of network latency, I do not use fine-grain decomposition. The machines I use, RS6K's, are so unevenly balanced, I must use a queue. This conflicts with the latency (request-response time) and, inorder to see appreciable speedup, I must initialize with 2 jobs per machine. This allows "asynchronous messages". J. Brian Grant ------------------------------------------------------------------------------- From: djc@dsmail.hmi.de (Markus Dahm) Hi, > Are you using PVM today? Probably :) I'm working on an comparison report between EXPRESS and PVM; and as far as I can see PVM wins the race. > What are your application(s)? Mainly we are trying to parallelize an FORTRAN(Yuk) application calculating cluster configurations of gas atoms. > How many and what kind of machines have you used with PVM? 9 SPARCS, 2 VAXes and a 4 processor CONVEX. I am quite fond of the PVM programming environment, it even may be that I will write my graduation using PVM. Yours sincerely --- __ ______ /\ /\ /\ \ | ___ \ Markus Dahm /\ \/\ \ / \ \ | | | \ \ / \/ \ \ / /\ \ \ | | | | ) Voice : +49 (0)30 805 61 34 / /\ /\ \ \ / / /\ \ \ | | | | | Office : +49 (0)30 8009 - 2594 / / /\/_/\ \ \ / / /--\ \ \ | | |__/ ) E-Mail : markusd@cs.tu-berlin.de / / / \ \ \/ / /----\ \ \ | |/__/ / news@dsmail.hmi.de /_/_/ \_\/_/_/ \_\_\|______/ djc@dsun3.hmi.de ------------------------------------------------------------------------------- From: Tom Bartol Hi, I use pvm 3.1 a lot. I'd use it more if the folks over here at the San Diego Supercomputer Center could get it working properly on their Paragon. I've written a Monte Carlo computer simulation of synaptic function in the nervous system. I've use pvm on a hetergeneous network of sparcs, SGI's, and RS/6000's. I'd like to be able to use it on the Paragon machine here but they've been having trouble getting it working. I really like pvm. Thanks for your wonderful efforts on this product. It's a great alternative to EXPRESS. Thanx again, Tom ------------------------------------------------------------------------------- From: Kensuke Shirakawa Q. Are you using PVM today? A. Yes, I am. Q. What are your application(s)? A. I am currently testing PVM 3.1, to see the relationship between the number of hosts in the virtual machine and the speed of a distributed matrix multiplier program. Q. How many and what kind of machines have you used with PVM? A. I have been using up to 8 DEC 5000 workstations connected by an internet. ------------------------------------------------------------------------------- From: louis@arthur.ee.up.ac.za (Louis Coetzee) Are you using PVM today? Yes. What are your application(s)? Parallel image processing algorithms. How many and what kind of machines have you used with PVM? IBM RS6000: 1*340 3*220 A pleasure. Louis Coetzee University of Pretoria, Pretoria, South-Africa ------------------------------------------------------------------------------- From: weimar@eole.ulb.ac.be (Joerg Richard Weimar) Are you using PVM today? yes. Also yesterday. In fact, I use it every day since it has become my standard interface between calculation and display. What are your application(s)? Cellular Automata, lattice gases. How many and what kind of machines have you used with PVM? Workstations: 1-4 Type: NEXT, SGI, SONY, SUN Any comments and suggestions for improvement are welcome. Thanks a lot for this tool. One thing that I am waiting for is the implementation of the InPlace data type (maybe it is already done, I haven't upgraded for some time (still 3.1 no patches)). One problem I have with upgrading is that the modifications I made to directories to support one of our workstations (add machine type SONY), got destroyed when I upgraded from 2.4 to 3.1. Since everything is running somewhat stable right now, I am reluctant to do that work again. -- Joerg R. Weimar #!###!!!##!!#!#!!!###!!##!!##!!!###!!!##!!##!#!#!#####!#!##! jweimar@ulb.ac.be (NeXTMail welcome) Service de Chimie Physique II, Universite Libre de Bruxelles #!##!##!#!#!#!!!##!!#!#!#!#!##!##!#!##!#!##!##!!#!#!#!##!#!# ------------------------------------------------------------------------------- From: paia@chert.CS.ORST.EDU (Akash L PAI) > Are you using PVM today? YES > What are your application(s)? NUMERICAL ANALYSIS, TOOLS FOR PVM > How many and what kind of machines have you used with PVM? USED PVM ON RS6000, TRYING TO INSTALL IT ON THE i860 ------------------------------------------------------------------------------- From: jjv5@retina.chem.psu.edu (Jim Vincent) Are you using PVM today? yes What are your application(s)? parallel implementation of a molecular mechanic/dynamics package written by someone else (in Fortran), that is widely used How many and what kind of machines have you used with PVM? SGI Indigo, 4D IBM RS6000 IBM SP1 Sun 4 series Suggestions: To get feedback for the sake of improvement how about asking what was/is the most difficult part about using PVM? My answer: Installing and using on many machines. It isn't that difficult but it is a pain in the but to copy and install over and over again. I have made self install scripts for installing on reomte machines from a master and also makefiles that copy source code and compile on all machines in a hostlist from a master machine. This greatly simplifies using PVM on a wide scale. ------------------------------------------------------------------------------- From: spector@gandalf.Berkeley.EDU (Phil Spector) >> Are you using PVM today? Yes >> What are your application(s)? Parallelization of the CART (Classification and Regression Trees) algorithm. >> How many and what kind of machines have you used with PVM? A network of 40 Sun Microsystems SparcStation 1+ 's. - Phil Spector Statistical Computing Facility Department of Statistics UC Berkeley spector@stat.berkeley.edu ------------------------------------------------------------------------------- From: kbarnes@hannibal.atl.ge.com (Kerry B Barnes) >Are you using PVM today? Yes >What are your application(s)? We are using it as the communication layer for our Graphical Entry Distributed Application Environment (GEDAE) that allows a user to build applications as data flow graphs and them map those flowgraphs around a heterogeneous network of processors. PVM was used to reimplement the communicaton layer of GEDAE giving us the benefit of having the system dependent communication code being automatically ported to a large number of processor types. How many and what kind of machines have you used with PVM? We have only used PVM on sparkstations, though we have had GEDAE (with the old communication layer) working on Vaxes running Ultrix and VMS, and on a several other processors. Being able to have flow graphs working in a heterogenous environment is important to us. Any comments and suggestions for improvement are welcome. Would very much like a multithreaded version of PVM, perhaps based on POSIX standard. This would allow us to avoid some of the complex workarounds we need to do to get asynchrony using unix signals, but without a "signal safe" unix library. In fact though we currently have this working reliably in our old communication layer (tested quite exhaustively over a period of years) we had to avoid using unix interrupts with PVM. Therefore though the current version of PVM gives us the portability we desire it does not allow us to provide the responsiveness availble in our current layer where processes can send each other preemptive commands. ------------------------------------------------------------------------------- From: Mike Norman I am currently not using PVM. I intendt o use PVM in the near future for a Vehicle Routing Application. Hardware is still undefined. MIke ------------------------------------------------------------------------------- From: Victor Eijkhout 1/ Yes I use pvm regularly 2/ Sparse iterative matrix solvers 3/ At most 8 hosts, Sun4, HPPA, RS6K. Victor Eijkhout Department of Computer Science; University of Tennessee at Knoxville 107 Ayres Hall; 1403 Circle Dr.; Knoxville TN 37996-1301; USA phone: +1 615 974 8298 (secretary 8295; fax 8296); home +1 615 558 3069 Support the League for Programming Freedom! lpf@uunet.uu.net ------------------------------------------------------------------------------- From: Frank.Pinski@UC.Edu (Frank Pinski) Yes. Some test problems. 15 Sun SPARCS Question: Is possible to spawn jobs with a nice value? (The procedure in the manual doesn't seem to work.) Frank Pinski ------------------------------------------------------------------------------- From: petri@ibr.cs.tu-bs.de (Stefan Petri) We have installed it at our institute but are rarely using it. Survey> What are your application(s)? We are investigationg load balancing in distributed systems, and look at pvm because it is a system that the engineering faculties here make heavy use of. Survey> How many and what kind of machines have you used with PVM? 25 Sun Sparc Stefan ------------------------------------------------------------------------------- From: merlo@vega.EPM.ORNL.GOV (Alessandro Merlo) Nowadays I am not using PVM. I used it a little bit last year connecting together an nCUBE with 16 procesors and a network of workstations (SUN3/50, SUN SPARCstations, DIGITAL DECstations). The applications I tested were matrix multiplication algorithms. Since I am developing a tool for workload characterization of parallel systems, and this toll will be able to analyze tracefiles collected using the PICL libraries (among others), I think I will use PVM & HENCE again in the next future. Best regards, Alessandro Merlo Universita' di Pavia Dipartimento di Informatica e Sistemistica Via Abbiategrasso, 209 I-27100 PAVIA ------------------------------------------------------------------------------- From: Roger M. Firestone We are not yet ready to use PVM; it has just been ported over to our operating center and has yet to be compiled. There are aspects of the documentation that are somewhat less transparent than others. I am perusing comp.parallel.pvm to learn about others' experience before plunging in. A possible first application will be for an e&m code for which portability of the resulting software will be a major consideration. We'll probably be using Intel, TMC, Sun, and NeXT equipment with PVM. Sorry I can't be more specific yet. Roger M. Firestone ------------------------------------------------------------------------------- From: Philippe Meunier > Are you using PVM today? No, i am using p4 because i need shared memory message passing on a KSR machine. I was told PVM will soon be able to use KSR's shared memory, so if i have time i will perhaps try to use it. > What are your application(s)? global optimization applied to chemistry. > How many and what kind of machines have you used with PVM? A network of 20 Sun machines. > Any comments and suggestions for improvement are welcome. Some asynchroneous message passing features would be usefull (something like the hrecv() function on the i860 machine). Philippe ----------------------------------------------------------------------- Philippe Meunier meunierp@anchor.cs.colorado.edu ------------------------------------------------------------------------------- From: sienicki@allegra.att.com (Jim Sienicki) Yes, I am using PVM today. I am running an automatic test-pattern generation program on several SUN workstations, using the fault parallelism paradigm. I have run the program on up to eight processors. No problems so far. Jim Sienicki ------------------------------------------------------------------------------- From: wka@cs.duke.edu (Bill Allard) I am not using PVM today. I was using it quite a bit two months ago. When I used it I programmed elementary linear algebra, such as matrix multi- plication, in order to get used to message passing and parallelism; PVM was very helpful in this regard. I used PVM on 25 sun4 workstations in the Duke University Mathematics Department. Bill Allard ------------------------------------------------------------------------------- From: mar@sg6.chem.upenn.edu (Wen Mar) I am using PVM today. My application is Molecular dynamics simulation. I typically use 3-4 machines in my project. They are IBM RS6000, SGI Indigoes, Kendall Square multiprocessor machines. Wen Mar Dept. of Chemistry Univ. of Pennsylvania ------------------------------------------------------------------------------- From: hsu@crl.dec.com Are you using PVM today? Yes What are your application(s)? Volume Rendering How many and what kind of machines have you used with PVM? 4-8 machines. DECstations, soon Alphas. ------------------------------------------------------------------------------- From: jjv5@retina.chem.psu.edu (Jim Vincent) Are you using PVM today? yes What are your application(s)? Parallel implementation of molecular modelling package How many and what kind of machines have you used with PVM? SGI Indigo, 4D IBM RS6000 IBM SP1 Sun 4 series ------------------------------------------------------------------------------- From: wentoh@CS.UCLA.EDU (Wen-Toh Liao) > Are you using PVM today? Yes. Well, sort of. At least not everyday. > What are your application(s)? I am developing and distributing a distributed simulation language that uses PVM as the lower level network communication support. > How many and what kind of machines have you used with PVM? Mainly, SUN4, SPARC, and DEC-station. wentoh ------------------------------------------------------------------------------- From: Soeren Olesen > Are you using PVM today? Occasionally. > What are your application(s)? Simulated annealing, image restoring. > How many and what kind of machines have you used with PVM? 1-32 SUN4 ------------------------------------------------------------------------------- From: V Varadarajan Are you using PVM today? Not today. But I am doing some developement using PVM 3.1. We are exploring its usefulness in distributed computing in electromagnetics. What are your application(s)? Mainly applied numerical analysis. Solution of iterative methods in linear systems and time evolution problems for large problem sizes. How many and what kind of machines have you used with PVM? DEC, Sparc and HP workstations. A few other comments: PVM calls increase the code size tremendously. I think you could provide some routines that can do all the work of pvm_initsend, pvm_pk* and pvm_mcast or pvm_send in one call to it. This will greatly help us keep the code size under control. You should also make provisions for symmetric multiprocessing in shared memory environment; its time will eventually come, perhaps soon. Rajan. u6532@c.nersc.gov ------------------------------------------------------------------------------- From: William T. Rankin Dear PVM Project Team. Firstoff, I'd like to thank you for a job well done with PVM. Here is you survey response. > Are you using PVM today? As we speak. > What are your application(s)? I am using PVM to compute a large numerical sequence called "Golomb Rulers". This forms the basis for my Masters Thesis work, which should be completed in a couple months. The total time required for this computation has been estimated in the ten's of thousands of CPU hours. The processing model I am using a a basic master/slave configuration, using a "processor pool" strategy for load balancing. > How many and what kind of machines have you used with PVM? Currently I am running in a semi-homogeneous network of various models of DecStations (from 2100's through 5000 series machines). At one point I was utilizing 92 processors. I have also run on a small network of 12 Sparc Classics under Solaris 2.2. > Any comments and suggestions for improvement are welcome. I have some comments and suggestions on the Solaris port, which I will post in the near future. thanks again, -bill rankin Duke University wrankin@ee.duke.edu ------------------------------------------------------------------------------- From: Michel Proulx > Are you using PVM today? Yes, we are. We are still in the development phase. > > What are your application(s)? > We are trying to parallelize the EGS (Electron Gamma Shower) code -- Monte-Carlo calculations -- with minimal modification of the existing code (Fortran), with dynamic load balancing, automatic task/host failure recovery, and dynamic addition and removal of participating hosts. > How many and what kind of machines have you used with PVM? > For now, only Sun workstations, but Silicon Graphics machines will be included soon. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. The inclusion of the "pvm group" task in the pvm daemon would be nice. It would give better robustness to the pvm virtual machine. (since the "group" task can be started on a different machine than the master pvm, if either machine dies, the pvm task are either killed or hang...) -- Michel Proulx, National Research Council Canada, Ottawa, Ontario ------------------------------------------------------------------------------- From: derek@crl.dec.com 1. Are you using PVM today? Yes. 2. What are your application(s)? We are using PVM to start up processes on a farm of workstations to act as a multiprocessor. PVM is also used to pass initial data and collect final data. 3. How many and what kind of machines have you used with PVM? 4 machines or few are currently being used with PVM. They are all MIPS-based DEC workstations at the moment. Derek Chiou ------------------------------------------------------------------------------- From: vavasis@cs.cornell.edu (Stephen Vavasis) (1) I used PVM last semester (1/93-5/93). (2) My application was educational -- I was teaching a course on parallel scientific computing. (3) We used PVM on RS6000's -- generally a cluster of eight. -- Steve Vavasis (vavasis@cs.cornell.edu) ------------------------------------------------------------------------------- From: dave@meaddata.com (Dave Pavey) We are currently trying to understand in a client/server/distributed object design, the impact of various couplings between cooperating systems. In effect the difference loose, tight, or as in the case of PVM distributed coupling. We are not currently running PVM, but are rather studying it from the perspective the effect of SMP vs loose coupling and RPC vs Distributed Objects on such a distributed coupling system. This understanding will form a basis for evaluating hardware selections in the future. Our commitment is firmly entrenched with the OSF/DCE,DME technologies and so there is a problem introducing PVM into our architecture. Where technologies are not required to be DCE implementations, they are required to interoperate with such DCE technologies as makes sense. The PVM distribute compute model has significant appeal, but unitl it can be integrated into the DCE environment we won't be able to do anything but investigative inquiry. Dave Pavey Sr. Staff Sys. Eng. Mead Data Central ------------------------------------------------------------------------------- From: rodney@oce.orst.edu (Rodney James) > Are you using PVM today? Yes. > What are your application(s)? 1. Distributed Monte Carlo evaluation of Feynman path integrals. 2. Ocean/atmosphere modeling on workstation cluster 3. Global atmosphere modeling on coupled parallel supercomputer and workstation cluster. > How many and what kind of machines have you used with PVM? RS/6000, HP9000, SUN Sparc, CM-5, CM-200. As many as 32 machines used at once, mostly RS/6000 and SUN4's. > Any comments and suggestions for improvement are welcome. I am very happy with pvm. No complaints, keep up the good work! -- -Rodney James (rodney@oce.orst.edu) Oceanography / Oregon State University ------------------------------------------------------------------------------- From: shaffer@phyast.pitt.edu (C. David Shaffer) Are you using PVM today? Yes. What are your application(s)? I use PVM primarly for number crunching. Currently I have two large applications running under PVM. The first calculates "matrix elements" for inelastic scattering of photons from bound (atomic) electrons (Compton scattering). This calculation has migrated across various parallel platforms (including the BBN Butterfly at LLNL, and a CM5 which used to be at the Pittsburgh Supercomputing Center) but I hope to be able to stick with PVM. The majority of the time consumed by this code is in performing numerical integrals and computing Clebch-Gordon coefficients. It was parallelized at the matrix element level and uses a 'Slave-host' configuration for dynamically assigning a matrix element to be computed by each slave. The number and size of the messages passed are small so this application works well over a busy network. The second application is calculating elastic scattering of photons from bound electrons. This calculation is still being parallelized however we have a version that is parallelized at the level of the atomic subshell from which the photon is being scattered. How many and what kind of machines have you used with PVM? I use 10 workstations including Sun SPARCstation 10, 2, ELC and HP 9000/730. I have access to a few MPP machines (including a CM5 now with NSC) which I hope to integrate into my 'virtual machine'. I have spent the summer at LLNL (in Livermore CA) and am trying to convince many of my peers that using PVM in parallel software development would represent a step forward, away from the machine dependent coding of the past, for our applications. Although the computing needs at the Lab may vary from division to division, here in L-Division, many applications can be parallelized with minimum changes and at a sufficient granularity as to keep message passing at a minimum. I think that such applications are well suited for PVM and hope to convince others. You have provided an excellent package for parallel programming which I hope will save much needed gov't time and money. Keep up the good work! ************************************* The opions expressed here are my own. I do not speak for Lawrence Livermore National Lab or the DOE. ************************************* ------------------------------------------------------------------------------- From: K P Wang > Are you using PVM today? Yes, I am using PVM. > What are your application(s)? Parallel numerical computation: parallel finite difference, parallel finite element, domain decomposition, and load balancing schemes. > How many and what kind of machines have you used with PVM? I am using clusters of SUN4 and SUN3. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. 1. I wish I could have of parallel I/O functions similar to the I/O in Express such that my read and write can be performed through the same consol. 2. I always have a hard time to understand the error messages and don't known what happen to my program. If there is a list of error messages and their cure then my life will be easier. 3. Thank you too. Kai-Ping Wang ------------------------------------------------------------------------------- From: Chas J Williams III 1. yes 2. molecular dynamics, weather 3. 15 machine rs/6000 cluster 7 machine cluster (3 hp 9000, 1 sgi, 1 sparc 10, 1 alpha, 1 rs6000 ) --chas iii chas@ra.nrl.navy.mil ------------------------------------------------------------------------------- From: Antoine Petitet Dear PVM Project Team, No, I am not using PVM today. My application basically uses PVM to run a program on a remote machine, sending the results back to my workstation. The remote machine is for the moment typically an Intel iPSC/2, iPSC/i860 or a IBM RS6K. It should include the CM-5 in the near future. Remark, it is difficult to install PVM on the I860 for example. Moreover, I didn't see any difference in the documentation and the makefiles to install PVM on the host machine of the hypercube (a PC 386) and the nodes (I860) themselves. In this particular case, I understand that the librairies for the host and the nodes should be stored in distinct directories, which is unclear in the documentation I got from xnetlib. (PVM vers. 3.1). Antoine Petitet ------------------------------------------------------------------------------- From: Hanebutte Ulf Rainer Are you using PVM today: Currently not for everyday simulations, the use of APPL has shown to give a more robust distributed environment. But keep PVM2.4 and 3.1 current. What are your application(s)? Aeroscience Applications, in particular adaptive mesh refinement algorithms for complex shock wave phenomena. How many and what kind of machines have you used with PVM? 16 SUN elc's 8 SUN Sparc 10 or 3 SGI normally as a homogenious cluster ------------------------------------------------------------------------------- From: haque@rcg.cc.iastate.edu 1. Yes, I will try to use it today. 2. Benchmarking. 3. Comments: Wish it had some sensible instructions for installing and startup. ------------------------------------------------------------------------------- From: Vasudha Govindan Are you using PVM today? No What are your application(s)? N-Body simulations Soln of Laplace eqn programs to measure ommunication times, etc. How many and what kind of machines have you used with PVM? 17 SUN workstations (SUN4s, Sparc-1,2,10) (full configuration) I normally use upto 8 workstations to test/debug programs. ------------------------------------------------------------------------------- From: "E. Loren Buhle, Jr. [(215)-662-3084]" >Are you using PVM today? Yes. > >What are your application(s)? 3 dimensional radiation therapy simulation calculations for photons and electrons. Some 3D medical imaging . . . >How many and what kind of machines have you used with PVM? I use Stardent (formerly Ardent), NeXt, DECstation 5000/240, 5000/200, and a DEC AXP/OSF 3000-500. There are two 5000/240s and one 5000/200. >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. I'm just getting started, but appreciate your survey and would be curious to see the results.... -- Dr. E. Loren Buhle, Jr. INTERNET: BUHLE@XRT.UPENN.EDU University of Pennsylvania School of Medicine Phone: 215-662-3084 Rm 440A, 3401 Walnut St., Philadelphia, PA 19104-6228 FAX: 215-349-5978 ------------------------------------------------------------------------------- From: Jeff Johnson > Are you using PVM today? > Yes, we are using PVM, but not in full production. Our primary application is still being tested and certified. > What are your application(s)? > Our major application is an in-house CFD code > How many and what kind of machines have you used with PVM? > Hewlett Packard 9000-715,720,730,735 and 755. Currently 7, potentially over 200 within the next 6 months. Silicon Graphics Crimson (1), Indigo (7) IBM RS/6000-320 (1). > Any comments and suggestions for improvement are welcome. Getting a 'pvm' started after a previous failure can be a real pain to an end user. If a previous 'pvm' is still running or if the 'pvmd.uid' file is still laying around, manual intervention is required. This is not acceptable if it is a batch job that is being started in the middle of the night!. A couple of things could help the situation. 1) Create a mode that doesn't require daemons on the remote machines. This would cause slower startup of tasks, but in our case this is not a problem since we start a task and it runs for hours. The 'p4' library from Argonne defaults to 'no-daemon' mode. Note that this eliminates the problems with undeleted 'pvmd.uid' files and old 'pvmds'. 2) Better checks for stray 'pvmd.uid' files. It probaby isn't necessary to bomb just because a 'pvmd.uid' file exists when trying to start 'pvmd'. If a 'pvmd.uid' file exists, try to connect to the pvmd. If the connect is unsuccesful, one can probably assume that the pvmd isn't really running. 3) The master application can't tell when the master 'pvmd' has started. We start 'pvmd' externally to the application with a host file. We had to put a 'sleep' command after the 'pvmd' command to allow it to start before trying to start the application so the application could successfully connect. The time to get started varied widely depending on system load. I mailed a message to the developers, who responded that I could use 'the start_pvmd(?) call from the application. This function is not documented and is not available from FORTRAN (see below). Also, we need full capability from FORTRAN (some very useful functions are only available from C). In particular, 'pvmfconfig' and 'pvmfspawn'. Also, another nasty...ALL externally visible names should have a name that starts with 'pvm' to avoid name space conflicts. Someday, somewhere, somebody will have to link with the pvm library and some other third party library which they have no control over and will get duplicate symbols! Not to sound too critical, 'pvm' has made my life a lot easier. The documentation is good (I've only had to look at the code a couple of times), and responses to mail messages to the developers have been very prompt! Hope this has been constructive! -- Jeffrey A. Johnson Phone: (314) 232-2248 McDonnell Douglas Aerospace-East FAX : (314) 777-1328 P.O. Box 516 Email: m191713@svmstr07.mdc.com MC 1064126 St. Louis, MO 63166 ------------------------------------------------------------------------------- From: pereira@alta.research.att.com (Fernando Pereira) > Are you using PVM today? No. I tried it briefly a while ago, but I have not had the time to try it out yet for the project I have in mind. > What are your application(s)? Training statistical models of natural-language text. The training technique involves partitioning a large training set among processors, having each build improved parameter estimates from its training material, averaging the resulting parameter estimates and reiterating. There is a natural way of organizing this computation for PVM, and in fact Yves Schabes at Mitsubishi Research Labs in Cambridge (schabes@merl.com) with whom I have collaborated has used PVM in exactly this way. > How many and what kind of machines have you used with PVM? A few SGI R4000 Indigos . Fernando Pereira 2D-447, AT&T Bell Laboratories 600 Mountain Ave, PO Box 636 Murray Hill, NJ 07974-0636 pereira@research.att.com ------------------------------------------------------------------------------- From: fatoohi@nas.nasa.gov (Raad A. Fatoohi) Are you using PVM today? Yes What are your application(s)? NAS Parallel Benchmark (CFD applications). How many and what kind of machines have you used with PVM? About 40 SGI workstations. Thanks! Rod Fatoohi, Ph.D. Mail Stop T045-1 NASA Ames Research Center Moffett Field, California 94035 Phone: (415) 604-3486 FAX: (415) 604-3957 E-mail: fatoohi@nas.nasa.gov ------------------------------------------------------------------------------- From: wieland@bobafet.nrl.navy.mil (Fred Wieland) Here's an answer to some of your PVM questions on the survey. Are you using PVM today? I have a program which uses multiple processes on a single machine. I have used PVM to distribute the processes to multiple machines, and it worked successfully. I demoed the resulting product, and it received only lukewarm reaction from the community. I have since had little or no time to continue working on it, although it is definitely an active product as far as I'm concerned. What are your applications? The application area is physics-based modeling and simulation. How many and what kind of machines have you used with PVM? I have run PVM with the processes mentioned above running on an SGI Indigo, and SGI Crimson, and an SGI personal IRIS. I have used at most 3 machines at any one time, because I've only configured 3 of the 12 processes with PVM. I plan to configure all 12 sometime in the future, although I don't know when. I hope this information helps. -Fred Wieland, NRL code 7604, Washington, DC ------------------------------------------------------------------------------- From: Robert P. Goddard PVM Survey, No, I am not using PVM today. My potential future PVM application is in simulation of underwater sound. The current version is the Sonar Simulation Toolset. It produces sampled sound as "heard" by a user-specified moving multi-channel active or passive sonar system in a specified ocean containing any number of moving targets (sound reflectors) and sound sources, plus reverberation, ambient noise, and self-noise. The resulting sound is suitable for listening, or for feeding to the back end of a sonar processor (or a computer model of one). It is used for testing and developing sonar systems, training sonar operators, predicting performance, developing tactics, planning experiments, and interpreting measurements. Reference (out of date): Proc. Oceans 89, IEEE Publication 89CH2780-5, pp. 1217-1222. I anticipate using a mix of a dozen or so Sun-4 and NeXT computers, and whatever else is available at that uncertain future date, if and when I use PVM at all. I downloaded PVM 3.0 from netlib@ornl.gov on 27 April 1993, after reading Dongarra et al., "Integrated PVM Framework Supports Heterogeneous Network Computing", Computers In Physics 7 #2, Mar/Apr 1993, 166-175. My motivation was mostly curiosity, to find out whether PVM has the potential to help me satisfy my very substantial computational requirements. Since then, I haven't had time to pursue the question further. Dr. Robert P. Goddard, Senior Physicist Applied Physics Laboratory University of Washington 1013 NE 40th Street Seattle, WA 98105-6698 206-543-1267 Internet: rpg@apl.washington.edu ------------------------------------------------------------------------------- From: Walter Scott Are you using PVM today? Yes, we still use it, although I am not sure today that the choice to use PVM was a wise one. We have experience with both PVM 2.4 and PVM 3.0 patch 1-2. We are primarily interested in hooking up HOMOGENEOUS machines, and so the immense overhead of packing/unpacking isn't worth paying for. I have rewritten our communications so as to use sockets, and get much better performance. What are your application(s)? We ( Florian Mueller-Plathe and myself Walter Scott) have developed a parallel Molecular Dynamics code written in FORTRAN-77. Parallel MD on workstation clusters has been difficult to achieve -- indeed, I would say has not been efficiently achieved to date-- because of the relatively high communication/calculation ratio. The ration, btw, is MUCH better for quantum chemical problems, and correspondingly many quantum code have been successfully parallelized. We have come up with methods which we hope will aleviate the communication problem ( publications in preparation). Quite a few MPP vendors promise to offer a PVM port to their HW, and so, hopefully we would be able to use Iour code there as well. See below for a comment on this. How many and what kind of machines have you used with PVM? 1) up to 20 sun WS IPXs connected by Ethernet. PVM 2.4 worked well, PVM 3.0 is slower! 2) 3 IBM 6000 WS connected by Ethernet and SOCC. Used a highly unsatisfactory PVM version (called PVMe) for the SOCC connection: the pvm abstraction is not supported properly: the programmer has to watch out for dead locks( buffer allocation problems), bcast or mcast NOT supported,.... Also used the Ethernet connections with pvm 2.4. This worked, but of course the comm. bandwidth is tto small. 3) On a 64 processor KSR1. The PVM 3.0 version would not compile properly. The release was faulty in that the c routine identifiers were missing a "_". When I did get it going, performance was DISMAL. 4)port of our code to Intel Paragon by intel people. They replaced our PVM 2.4 communications with intel specific ones. IN comparing the the versions, the PVM version is much slower. From all of this I conclude: use PVM for portability ( but even then some vendors use PVM 2.4 and others use PVM 3.0 -- as far as I can see, the changes in user interface are cosmetic, provide few if any improvements and are simply a p. in the a.) BUT NOT FOR PERFORMANCE. This is true for WS clusters and even more so for MPPs. I have no experience using your graphic dooh-dahs, as I don't need them. Consider the following: the single most important reason for going parallel is the quest for performance. As far as I am concerned, PVM has yet to deliver. with kind regards, Walter Scott, Laboratory of Physical Chemistry, ETH Zuerich, Switzerland Tel. 0041 (1) 256 55 03 Fax. 0041 (1) 252 34 02 ------------------------------------------------------------------------------- From: Pedro A.M. Vazquez > Are you using PVM today? Yes, > What are your application(s)? a) Quantum chemistry calculations b) Molecular Graphics applications > How many and what kind of machines have you used with PVM? Versions 2.4.1 and 3.1 on AIX, SunOS4.1.1-2-3 and 386bsd0.1 (Jolitz) AIX: two 320h and 8 model 560 SunOS: a mix of 8 sparcstatios1+, ELCs , 330 and 390 servers 386bsd0.1: 386/387 40MHz pc clone, 486 33MHz pc clone > Any comments and suggestions for improvement are welcome. Sugestion: to create a mail exploder to redistribute comp.parallel.pvm to the people without news access. Now we have set up a listserver to do this in Brasil but it seems these people can't post through cs.utexas.edu to c.p.p. Sugestion: what about a pvm examples/contrib repository? -- ============================================================================= Pedro A. M. Vazquez | Caixa Postal 6154 | FONE: 55 192 39 7253 Depto de Fisico Quimica | CEP 13081 | FAX : 55 192 39 3805 Instituto de Quimica | SP/BRASIL | UNICAMP | |vazquez@iqm.unicamp.br ============================================================================= ------------------------------------------------------------------------------- From: zeitler@trantor.si.com (Dave Zeitler) > Are you using PVM today? No, but will be shortly. > What are your application(s)? Empirical tests of parallel random number generation using S-Plus running concurrently on a network of sun stations and/or an Ncube II. > How many and what kind of machines have you used with PVM? I've run up to 12 Sparc's at once before while working with PVM 2.4. The network I'll be working on now is 18+ Sparc's ranging from Sparc-1 to Sparc-10's. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. Not right now, but I'll likely have some comments after I get the random number stuff running and start running the testing. -- Dave Zeitler zeitler@si.com Smith's Industries Ph: (616)241-8168 Aerospace & Defense Systems Inc. Fax: (616)241-7533 ------------------------------------------------------------------------------- From: Ludek Matyska > Are you using PVM today? I am not using PVM directly, but my PhD student (one) is working with HeNCe and PVM (and will definitevely work with after holidays). > > What are your application(s)? The computational chemistry, e.g., flexibility analysis (a paper about using PVM was sent to J. Comp. Chem. but, alas, I don't know its current state), molecular modelling (we would like to port Allinger's MM3 to parallel (distributed) environment using the PVM). > > How many and what kind of machines have you used with PVM? The largest VM used consisted of 8 Sparc's 1+ plus 4 DECStations 5000/120. We have some preliminary results using BSDI and SCO UNIX on Intel 80486 based boxes. Best luck with the PVM development Dr Ludek Matyska UVT MU Brno School of Informatics Buresova 20 City University 602 00 Brno London Czech Republic UK e-mail: ludek@muni.cz ------------------------------------------------------------------------------- From: wangy@gmsis.ms.ornl.gov (Yang Wang) > Are you using PVM today? Not today. But, will be using it tommorrow. > > What are your application(s)? ab initio electronic structure calculations. > > How many and what kind of machines have you used with PVM? 6 machines. DEC-stations, IBM-stations, iPSC-860, Paragon, etc... ------------------------------------------------------------------------------- From: Al Stone >Are you using PVM today? Yes, but on a sporadic basis. >What are your application(s)? I typically act more as a "help desk" than an end-user, due to the nature of my job. >How many and what kind of machines have you used with PVM? Up to 9 HP9000/735s on ethernet and/or FDDI; I have worked with others running 60+ machines (ranging from SPARC2's to other HPs to whatever they could get their hands on). More often than not, the usage has actually been on homogeneous clusters of machines. >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. Could you please indicate what some of the upcoming features might be? Or provide a mechanism for discussing them? Or, even provide a "beta" version? Thanks for asking! -- Ciao, al aka Al Stone Performance Consulting Team WSG Technical Consulting Workstation Systems Group (WSG) Hewlett-Packard Company 3404 E. Harmony Rd. Fort Collins, Colorado 80525 E-mail: al@hp_rigel.fc.hp.com Phone: 303-229-2399 Fax: 303-229-3002 ------------------------------------------------------------------------------- From: phanau@wv.MENTORG.COM (Paul Hanau) > Are you using PVM today? Yes. > What are your application(s)? Here at Mentor Graphics we develop EDA (Electronic Design Automation) software. I'm in a technology-oriented group responsible for shared "framework" technologies -- I work in the area of distributed computing support and intertool communications. I've been showing PVM to various end-user development groups to see if it might help them solve their problems in areas like distributed circuit simulation, etc., rather than re-inventing a similar system from scratch. > How many and what kind of machines have you used with PVM? I have PVM going on SUN4 and HPPA workstations thus far. I've been using the fractal display demo to illustrate the use of PVM, with up to 20 workstations involved, but 6-10 is more typical. > Any comments and suggestions for improvement are welcome. I've modified the demo to handle missing servers more gracefully and allow you to click on a band to identify which server handled it. This is useful in identifying slow-pokes (i.e. servers that take longer to do their piece than anyone else, in which case it might be faster to delete them from the configuration). If you'd like the modifications, let me know and I'll send them along. I've had no real problems with the software thus far. The documentation is very good (better than a lot of systems that cost money). If you have any references to uses of PVM in EDA I'd be interested in seeing them. Keep up the good work. ------------------------------------------------------------------------------- From: seyfarth@whale.st.usm.edu (Ray Seyfarth) > Are you using PVM today? Yes. > What are your application(s)? > Ocean modeling and parallel algorithms in general. > How many and what kind of machines have you used with PVM? > IBM RS6K 3 SGI 2 SUN4 3 ------------------------------------------------------------------------------- From: Dick Wilmoth > Are you using PVM today? Yes. > What are your application(s)? Direct Simulation Monte Carlo (flow simulations in rarefied gas dynamics) > How many and what kind of machines have you used with PVM? Sun Workstations (up to 32) SGI Workstation (4 processor) Cray YMP (8 processor) -Dick Wilmoth- ============================================================================ NASA Langley Research Center Email: wilmoth@ab00.larc.nasa.gov Mail Stop 366 Phone: (804) 864-4368 Hampton, VA 23681 ============================================================================ ------------------------------------------------------------------------------- From: sunny@cs.wm.edu (J. Sunny Egbo) |>Are you using PVM today? Yes. |>What are your application(s)? We are developing a set of C Library routines to facilitate running of Intel-like (Intel iPSC/2) commands within PVM. We are in the final stages of our testing. We are interested in running parallel simulations over PVM. So far everything is working like a charm except for the problems I have pointed out below. |>How many and what kind of machines have you used with PVM? About 11 IBM RS/6000 workstations. |>Any comments and suggestions for improvement are welcome. |>Please send the information to pvm-survey@cs.utk.edu. |>Your help is much appreciated. When we started using PVM, we experienced some sporadic hanging of the PVM daemons. We think that RS/6000 kills the largest process in the system if it runs out of paging space and hence PVM was getting killed. We have bumped up paging space on all our workstations and the problem seems to have gone away. Also, the timing of sends and receives over PVM is still unclear to us. A few months ago there were discussions on comp.parallel.pvm where someone posted a timing program (i believe it was called times3 or so). We ran this code between two of our workstations but were unable to get performance even close to what was reported by others. Any additional information about what performance numbers that we should expect given a network configuration will be appreciated. -Sunny ------------------------------------------------------------------------------- From: sfleisch@hydra.convex.com (Steve Fleischman) > Are you using PVM today? Yup. > > What are your application(s)? Molecular dynamics, neural networks. > > How many and what kind of machines have you used with PVM? HP clusters. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Fix the bugs in 3.1. Pvm_spawn, pvm_tasks, pvm_mytid. Also, please limit the anticipatory documentation. I.e., a feature is planned so the documentation reflects what the function will do when that feature is implemented not what is in the code currently. E.g., pvm_tasks and the where parameter. The "heuristic" used by pvm_spawn. The documentation should mention pvm_tidtohost. This routine is very useful. Otherwise, keep up the good work. ------------------------------------------------------------------------------- From: fsset@corelli.lerc.nasa.gov (Scott Townsend) Are you using PVM today? yes, 3.1 patchlevel 5 & 3.2mpp What are your application(s)? no real application yet, I'm primarily just trying to see what PVM does & how at this point. My current vehicle for these investigations are the NAS parallel benchmark codes BT, LU, and SP ported from the iPSC library versions. How many and what kind of machines have you used with PVM? Primarily 32 RS6000's in a cluster, also a 32-node iPSC/860 Any comments and suggestions for improvement are welcome. Warning: I used to be involved with the development of APPL, another message passing library. Most, if not all, comments below are basically derived from this biased viewpoint. (these are in no particular order) - My first surprise/struggle was with PVM's execution setup being quite different from what I expected in a number of aspects: - though I specify a directory for my remote executable, that executable's current directory is NOT set to the directory I specified, but rather my home directory on the remote machine. I can see why this comes about, but it's quite inconvenient. I'd much rather the pvmd put the default directory where the executable is found, at least if it's been specified. I know I could make the executable a script which does the cd, but then I have to specify the directory to both PVM and the script. - Interrupting a run with control-c isn't propagated to all processes. In general, I prefer any error exit by any process to bring down the other processes. (i.e. one process dumps core => other processes halt) I suppose this could be implemented with the notification stuff, but that's both a slight pain for the user and currently not possible since apparently the notification message doesn't carry the exit status of the process. - TIDs are a pain. Invariably I want to think in terms of process ID's runing from 0 to N-1. Thus I always have to map my ID to a TID. - The API requires too much work on the part of the user who's trying to get the optimal performance. For the simple & common case of sending an array of numbers, PVM should be able to make the decision about what encoding scheme to use, handle the initsend, pack, send stuff. I'm thankful for the low-level access to the alternatives, but most users I've seen don't want to hassle with all the alternatives and then either go PVMDEFAULT of PVMRAW everywhere, breaking the application in a heterogeneous environment or running unnecessarily slow. PvmDataInPlace would often help if it existed. I realize the problem for multicasting for doing optimal packing, but 90% of the time it's a send directed to a specific process. A higher level API could get optimal performance for 90% of the cases which are really quite simple. (simple things should be easy, complicated things should be possible) - There appears to be a significant performance hit on the iPSC. I haven't traced it down yet but I'm suspecting it's the XDR & buffer copying that's hurting us. Again, PvmDataInPlace, automagically selected by the PVM library, would be a real help. - There are a few applications here which find it useful to have multiple processes an a single RS6000. (multi-block CFD codes) PVM's performance within a processor, even using PvmDataRaw and PvmRouteDirect is poor. I'm getting about 2MB/sec for large messages and about 1msec latency for small messages. For comparison, I get 30 MB/sec thruput & 200 usec latency with APPL. The reason for this is the difference between the PVM socket/pipe vs. APPL's shared memory. I think it would be real useful if PVM used shared memory between processes within a workstation. From my APPL experience I know this isn't easy, especially the correct cleanup aspect, but the performance increase is significant. - This is probably my own error, but I haven't figured out what's wrong yet: our cluster of RS6000's has two seperate Ethernets along with an IBM ALLNODE switch. We can accesses these via IP names of lace01, lacemp01, and lacev701 respectively. It seems that I can only use the "native" lace01 name as the first name in the hostfile. Although I'm logged into lace01, which is the same machine as lacemp01, if the hostfile starts with lacemp01 I get various errors I don't understand quite yet. (lace01 is considered "native" in that that's what is returned by `hostname`) If there is some code internal to PVM which compares gethostname() with hostnames in the hostfile there could be a problem with machines supporting more than one network. - Online manual pages would be nice. With the AWF package you don't have to assume the site has DWB/nroff. - pvmd's idea of static load balancing is better than nothing, but doesn't allow for muti-user realities. For the cluster here we've built tools to grab a load average snapshot and assign processes on that basis. Not great, but better than assuming the machine is all one user's. APPL made building such tools straightforward through the concept of a process definition file. I've done something similar for my APPL over PVM library, using the SPMD example as a basis. It would be handy if either pvmd knew about ruptime/rstatd/etc or there was some way for a user to alter pvmd's concept of a host's load. (maybe this could be done with machine speeds somehow, I don't know) I'm trying to address some of the above issues for myself in APPL and iPSC compatibility libraries I'm building on top of PVM. This should do fine for me, but it's possible other PVM users out there have similar concerns with the current state of things. ------------------------------------------------------------------------------- From: sanjai@ctd.comsat.com (Sanjai Bhargava) pvm-survey> Are you using PVM today? No. But I have used it for projects in the past. ( About 6 months ago ). pvm-survey> What are your application(s)? I had used it in video coding, primarily to distribute the motion-estimation processing to different machines. pvm-survey> How many and what kind of machines have you used with PVM? I have used it on HP-9000 series machines ( 720's and 730's ), SGI Crimson VGXT and SUN-4. The maximum number of machines at any time were about 20. pvm-survey> Any comments and suggestions for improvement are welcome. Thanks for creating a useful application. The only drawback I felt were that there was no way to dynamically get the machines available. If any machine in the list was the down, the program would bomb. Also the graphical tools ( HENCE ) was not very useful and not well documented. Thanks, once again. sanjai -- Sanjai Bhargava COMSAT Labs. (301)-428-4502 ( phone ) 22300 Comsat Drive (301)-428-9287 ( FAX ) Clarksburg, MD 20871 e-mail: sanjai@ctd.comsat.com ------------------------------------------------------------------------------- From: Mohamed Faisal >Are you using PVM today? No. >What are your application(s)? The main work I did was to convert an existing program named SAP (By Dr. Raphael Finkel), which was originally written for Hypercube, Sun , Sequent etc , I rewrote it using PVM function. The goal was since PVM already takes care of the underlying architecture in SAP we dont need to worry about anything. The SAP which stands for Simple Application Package was written to help write parallel programs. >How many and what kind of machines have you used with PVM? Two. Sun workstations and Intel Hypercube. SUGGESTION: I felt the need for something like asyncronous receive (like intels hreceive) where you can pass a pointer to a function to the receive function and then do your own work. When there is any input , it should cause an interrupt and call the users function. ------------------------------------------------------------------------------- From: thomas@greisen.physics.utah.edu (Stanton Thomas) Are you using PVM today? Yes. What are your application(s)? I am using PVM to ray trace various optical detector configurations for use in monte carlo simulations of cosmic ray observatories. The ray tracing results are subsequently used via lookup tables in PVM implementations of the monte carlo. How many and what kind of machines have you used with PVM? I have used PVM on IBM RISC 6000 (RS6K), Decstation 3100 (PMAX), and on Silicon Graphics (SGI) machines. Any comments and suggestions for improvement are welcome. We have 14 Motorola 68030 based machines which are currently running the VxWorks operating system. VxWorks is a real time operating system with TCP/IP. I would like to be able to extend PVM to this system. We will have from 54 to 162 machines within the next three years. If anyone is working on a port for VxWorks we would be very interested. If not, do you have any notes/suggestions for developing PVM ports. Stan Thomas Cosmic Ray Research University of Utah (801) 581-8650 ------------------------------------------------------------------------------- From: kernst@jpmorgan.com (Ken Ernst) PVMers, I am currently using PVM 3.1. Genetic Algorithms and Monte Carlo simulation. Up to 20 nodes in a heterogeneous configuration of IBM RS/6000 and SUN SPARC2s, IPXs, SPARC10 model 41s. Have the pvm_config command return the correct number of architectures and indicate a model number. For example SUN4 IPX. And give us a way to set the SPEED parameter. Regards Ken P.S. Great product. ------------------------------------------------------------------------------- From: delbaeth@mule.fhcrc.org (Robert Christ) >Are you using PVM today? We currently have PVM running but it is not being used. We have plans to use it in the future but due to some problems in code we have tried to switch to PVM things are progressing slowly. >What are your application(s)? > >How many and what kind of machines have you used with PVM? Right now it is running on 5 RS6000s. We plan to include a large pool of SPARCstations in the near future. Rob Christ Unix Network Administrator Public Health Sciences Division Fred Hutchinson Cancer Research Center ------------------------------------------------------------------------------- From: roscoe@BUCRF15.BU.EDU (Roscoe Giles) From: Roscoe Giles Deputy Director, Center for Computational Science Boston University Boston MA, 02215 617-353-7082 roscoe@roscoe.bu.edu ARE YOU USING PVM TODAY? yes WHAT ARE YOUR APPLICATION(S)? (1) We have a newly developed set of courses on parallel scientific computing with a special emphasis on MPP. PVM is being used to develop example applications linking workstations to our CM-5. There is a little work using PVM as a model distributed parallel system. (2) Some of our researchers are using PVM for research applications similar to the teaching application (1) -- ie human interface and graphics on workstation/ linked to CM-5. (3) One group has used PVM and hence to link a workstation cluster on campus to one at a local High School which is linked to us. HOW MANY AND WHAT KIND OF MACHINES HAVE YOU USED WITH PVM? CM-5 (here, NCSA, TMC) SGI SUN RS6000 ------------------------------------------------------------------------------- From: Jari Toivanen > Are you using PVM today? Not at the moment. > What are your application(s)? Solving partial differential equations. > How many and what kind of machines have you used with PVM? Few HP and Sun workstations. Jari Toivanen (toivanen@jyu.fi) Memories are lost in time University of Jyvaskyla, Finland like tears in rain. ------------------------------------------------------------------------------- From: tadams@bach.NOSC.Mil (Thomas A. Adams) > Are you using PVM today? Yes. > What are your application(s)? I have implemented my own "standardized" general purpose master-slave computing software (broadcast, parallel computation, commutative-associative collapse repeated over and over). > How many and what kind of machines have you used with PVM? Primarily Sparcstation networks. I have done a little bit with Convex. Attempted to use SGI but without success. I expect to use the Paragon in a few months. > Any comments and suggestions for improvement are welcome. The user's guide (ORNL/TM-12187) is good, but it might be worthwhile to also have (perhaps in a separate document) more examples and guidelines for good PVM programming practice. A standard PVM test suite would be good, if you do not already have one. ------------------------------------------------------------------------------- From: swk@mlb.semi.harris.com (Song Koh) pvm> Are you using PVM today? Yes. pvm> What are your application(s)? General program to program communication. In the domain of semiconductor device modeling. Communication between C programs, Scheme Programs, Spice simulators, Splus statistics package plus a few others. pvm> How many and what kind of machines have you used with PVM? Mostly SUN SPARC. But also HP9000, and DEC MIPS. pvm> Any comments and suggestions for improvement are welcome. Good job! ------------------------------------------------------------------------ Song Koh Harris Semiconductor swk@mlb.semi.harris.com MS 62-022 Voice: (407)-724-7085 P. O Box 883 (US mail) Fax: (407)-729-4960 Palm Bay Road (UPS) Melbourne, FL 32902-0883 ------------------------------------------------------------------------ ------------------------------------------------------------------------------- From: lucier@math.purdue.edu (Brad Lucier) > > Are you using PVM today? Not yet. > > What are your application(s)? > Flows in porous media > How many and what kind of machines have you used with PVM? > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > > Best wishes, > The PVM Project Team > > Brad Lucier ------------------------------------------------------------------------------- From: "Dr. Eric C. Frey" > > Are you using PVM today? we are in the process of developing code that uses it. We don't have any production applications running yet that use it (a few test applications). > > What are your application(s)? > We are writing applications that do 3-D reconstruction of single photon emission computed tomography images including compensation for physical image degrading factors. This is essentially a matrix inversion problem on very large matrices (8192x8192) using very noisey data. > How many and what kind of machines have you used with PVM? > we have 8 decstations (5-5000/200's, 1-2100, 1-3100); 1 stardent 3020, 2 stardent 750 (3020 in small box). We are planning on buying 2 Alpha 400's. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Our application could benefit from use of multicasts to distribute data sets. Also, help in load balancing (or providing the information to do it) and/or fault recovery. Eric Frey ------------------------------------------------------------------------------- From: harkin@fubar.cs.montana.edu (Gary Harkin) } } Are you using PVM today? Yes. } } What are your application(s)? Distributed systems for performance improvement. Primarily statistical mechanics. } } How many and what kind of machines have you used with PVM? Mips-based Decstations, HP 9000's. ------------------------------------------------------------------------------- From: solberg@radonc.ucla.edu (Tim_Solberg) We are using PVM to improve the performance of Monte Carlo radiation transport calculations for medical physics applications. We are running the MCNP Monte Carlo code developed at Los Alamos NAtional Labooratory. Our present version uses PVM 2.4, but I understand that a version using PVM 3.1 is ready for beta testing. We should begin this here in the next 1 to 2 weeks. We currently run on several Suns (SPARC2s) within our department and several more in other departments on campus. We have two MP workstations (4 SPARC2 processors each), thus PVM 3.1 should ofer substantial performacnce improvement once the Sun MP platform is supported. MCNP/PVM also runs on RS/6000s. For more information on MCNP/PVM, contact Gregg McKinney (gwm@lanl.gov) at Los Alamos. Thank you for your continued efforts on PVM. Timothy D. Solberg UCLA School of Medicine Dept. of Radiation Oncology 200 UCLA Medical Plaza, #B265 Los Angeles, CA 90024-6951 solberg@radonc.ucla.edu 310-825-4323 ------------------------------------------------------------------------------- From: Alan Ferrenberg > > Are you using PVM today? If you mean "have I run PVM today, Monday August 16?" the answer is no. I've used it within the past week. > > What are your application(s)? Molecular Dynamics simulations of fiber growth Cluster-Hybrid Monte Carlo simulation algorithms > > How many and what kind of machines have you used with PVM? RISC/6000s (several configurations with up to 7 processors) 386 PCs running BSD/386 (2 PCs) up to 15 processors on an IBM SP1 ------------------------------------------------------------------------------- From: sverre@nrel.gov (Sverre Froyen) > > Are you using PVM today? Yes. > What are your application(s)? Electronic structure calculations for solids. > How many and what kind of machines have you used with PVM? Sun Sparcstations, IBM RS6000. > Any comments and suggestions for improvement are welcome. A global syncronization call (pauses until all processes have caught-up). -- Sverre Froyen (+1-303-231-1782; FAX: +1-303-231-1271) NREL (National Renewable Energy Laboratory), Golden, CO 80401 email: sverre@nrel.gov ------------------------------------------------------------------------------- From: dwhite@mrj.com (David White) !. 1. Yes, I'musing pvm. 2. It serves as the basis for a mesage 2. It serves as the message passing system for a distributed application. 3. We are using Sparcs and a CM5. Dave (dwhite@mrj.com) ------------------------------------------------------------------------------- From: mobarry@maxwell.gsfc.nasa.gov (Clark M Mobarry) Are you using PVM today? Yes What are your application(s)? Computational Fluid Dynamics How many and what kind of machines have you used with PVM? SGI Indigo's, MasPar front end (Decstation), YMP-EL ------------------------------------------------------------------ Clark Mobarry mobarry@maxwell.gsfc.nasa.gov NCCS | voice: (301) 286-2081 Code 934, NASA/Goddard Space Flight Center | FAX: (301) 286-1634 ------------------------------------------------------------------ ------------------------------------------------------------------------------- From: fred@network.ucsd.edu (Fred Criscuolo) Hi, Yes we are currently using PVM. Our primary apllication right now is to do concurrent searches of the Brookhaven Protein Database on a series of complex keyword searches. We are also hoping to develop a PVM version of a critical component of a computational chemistry code. We are using SGIs and RS/6000s. Fred fcriscuol@ucsd.edu ------------------------------------------------------------------------------- From: js9s@uvahey.phys.Virginia.EDU >>Are you using PVM today? No, but I used it yesterday. >>What are your application(s)? I use it for data processing. Basically tape io speed is about 6 times the proceessing program speed, so it's natrual try to use one tape io device and distributing the events to other machines. >>How many and what kind of machines have you used with PVM? I run it on DECstations and SGI. about 15 of them. >>Any comments and suggestions for improvement are welcome. I know some fellow outside the PVM project is trying to port the thing to VMS. That's a very useful stuff considering the huge left-overs from last 15 years. thanks .j.g. ------------------------------------------------------------------------------- From: rsimms@math.clemson.edu (Robert Simms) Just recently used PVM in my masters project. Implemented a linear (least squares) solver using Givens rotations to obtain a triangular form of a large matrix. Givens transformations were done in parallel using PVM. Large number of communications seemed to hang daemons, so I bundled transformations together in messages to cut down on number of messages sent, but still, if I use a large enough matrix the daemons time out trying to communicate with each other. Tried to ask around for help, but only got: "pvmd could get swamped with messages" I used Sparcstations running SunOS 4. Roughly 4 of them were at my disposal. - Rob ------------------------------------------------------------------------------- From: fshyun@borey.lerc.nasa.gov (Hyun Kim) > > Are you using PVM today? yes > > What are your application(s)? computational fluid dynamics > > How many and what kind of machines have you used with PVM? IBM RISC machines ------------------------------------------------------------------------------- From: Po Shan Cheah > > Are you using PVM today? Yes. > > What are your application(s)? Multivariate integration in a distributed processing environment. > How many and what kind of machines have you used with PVM? Only 4 Sun Sparcstations. +----------- | Po Shan Cheah Columbia University SEAS | | cheah@cs.columbia.edu New York, NY | pcheah@nyx.cs.du.edu -----------+ ------------------------------------------------------------------------------- From: duncan@mercury.chem.utah.edu (Thomas W. Duncan) At this moment, I am working on upgrading some of our software from pvm2.4 to pvm3.1.5. We are using it to solve transtition state theory problems in theoretical chemistry. Up to this point, we have been running all of our code on a cluster of RS6000's (4), but will probably be adding an SGI machines to this soon. I would just like to say that compilation of the source code on the RS6000's was VERY easy. The makefiles worked perfectly and made me very happy. Wendell Duncan Chemistry Dept. University of Utah ------------------------------------------------------------------------------- From: vieux@chief.ecn.uoknor.edu I have compiled pvm on an IBM RS6000 Model 520 but have not really used it yet. When running it would be on 2-RS6000 Model 220's and the 520. Baxter Vieux ------------------------------------------------------------------------------- From: Peter McGavin >Are you using PVM today? So far I've only run some of the examples. >What are your application(s)? Intend to parallelise existing meso-scale wind model (meteorological). >How many and what kind of machines have you used with PVM? 5 Sun SPARCstations (SUN4) - 2xSPARCstation 1, SPARCstation 2, ELC, SLC 2 HP workstations (HPPA) - HP9000/705, HP9000/730 1 DECserver 5400 (PMAX) - 5400 Regards, Peter McGavin. ------------------------------------------------------------------------------- From: Peter Geschiere > > Are you using PVM today? I started with some experiments. > What are your application(s)? A parallel Navier-Stokes solver. > > How many and what kind of machines have you used with PVM? A 4 processor IBM RISC System/6000. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. I am planning to use pvm version 2.4. I know that version 3 has more primitives and is supplied with a much better manual, but I also heard that this version still contains some bugs. Is this true, and if so, when do you expect to have these bugs solved? Regards, Peter Geschiere. ------------------------------------------------------------------------------- From: "Andrew J. Caird" Are you using PVM today? No, I am just learning it. What are your application(s)? Hopefully paralleization of neutron diffusion codes used in nuclear reactor analysis. How many and what kind of machines have you used with PVM? Have tried with IBM RS6k and HP 700 series. As above, I am a true beginner and thus have really nothing to show for it. Thank you. --Andy ------------------------------------------------------------------------------- From:davis@masig.fsu.edu IRIS workstations and our CRAY YMP. >Any comments and suggestions for improvement are welcome. The computing center has not upgraded to version 3, whereas I have done this for the application that runs on the workstations. The restriction that both server and client must run the same version is restrictive. It may be desireable for major changes, 2.4 to 3.0, but should be avoided for minor ones like 2.4.1 to 2.4.2. I have several workstations and many users on each and I would like to install a single version of pvm for all machines/users. This is rather difficult for the current version. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. > >Best wishes, >The PVM Project Team > _______________________________________________________________________________ Alan Davis | INTERNET davis@masig.fsu.edu Mesoscale Air-Sea Interaction Group | The Florida State University | Room 8A Love Bldg. | Phone 904-644-3798 Tallahassee, FL 32306-3041 | FAX 904-644-4841 _______________________________________________________________________________ ------------------------------------------------------------------------------- From: kern@masc5.rice.edu (Michel Kern) > > Are you using PVM today? Not today, Aug. 16th, but this week, probably yes. > > What are your application(s)? In a broad sense, seismic inversion. More precisely, we (repeatedly) solve PDE's (for instance, the wave equation) with multiple right-hand sides. We use PVM to distribute the RHS over different computers. > > How many and what kind of machines have you used with PVM? Not more than 8. Mostly SUN's (Sparc 1, 2, 10), IBM RS6000, and soon HP. Tried on Paragon, no luck. > > Any comments and suggestions for improvement are welcome. Thanks for the very useful tool, and the support. -- Michel Kern Department of Computational and Applied Mathematics Rice University, PO Box 1892, Houston Texas 77251-1892 Tel: (713) 527 8750 x2288 E-mail: kern@rice.edu ------------------------------------------------------------------------------- From: Andrew Lumsdaine Are you using PVM today? Yes. What are your application(s)? Semiconductor device simulation. Waveform methods. How many and what kind of machines have you used with PVM? o Cluster of eight RS/6000 workstations with Sparc-2 as host o Cluster of 14 Sparc-10 workstations Any comments and suggestions for improvement are welcome. I recently read a tech report by Grant and Skjellum entitled "The PVM Systems: An In-Depth Analysis and Documenting Study." (I assume the PVM Project Team is aware of this report as several members of the team were listed in the acknowledgments). There seemed to be a fairly long list of cons and some of them seemed to be quite serious. Will these be addressed future releases of PVM? -------------------------------------------------------------------- Andrew Lumsdaine Dept. Comp. Sci. & Engr. email: Andrew.Lumsdaine@nd.edu 353 Fitzpatrick Hall phone: (219) 631-8716 University of Notre Dame fax: (219) 631-8007 Notre Dame, IN 46556 -------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Oleg Svintsitski I am using PVM 2.x for parallel computing of the multistage stochastic programming problems. The idea is to split an original decision tree between machines by assigning its leaves to different processors with providing the information exchange between the root and leaves through PVM. I've been using up to 8 IBM RS/6000. The limitation of using IBM's only is caused by the availability of the OSL software that is used as a LP-solver in the algorithm. It seems that PVM (i.e. presence of communication between different machines) slows things down a lot. My research is intended to show the advantage of parallel computing for this kind of computations, but I am not getting too much advantage against just serial computing so far. Hopefully switching to PVM3.1 will give more flexibility and speed up the computations. If you have any suggestions about how to increase the communication speed, I would really appreciate them. Oleg Svintsitski. ------------------------------------------------------------------------------- From: skumar@cadcam.eng.ohio-state.edu (Sanjay Kumar) > Are you using PVM today? Yes. Very much. I am a grduate student and PVM is very forms the very core of my MS thesis related research. Well, its not PVM rather applications needing PVM. > What are your application(s)? Engineering applications : finite element analysis which reduces to large-scale equation solving and engineering design using genetic algorithm. > How many and what kind of machines have you used with PVM? 11 IBM RS/6000, CRAY, CM-2 (ethernet connection) I believe its a great effort. We are really fortunate to have such a portable system available in public domain. My suggestions: In the long run: I. There should be some way of visualizing the program performance. I am looking for something that can give us kiviat diagram and Feynman's diagram as seen in papers by Geist and Sunderam. II. Portable and integrated port to multiprocessors like CM-5 which will allow us to bypass the complicated CMMD and other such vendor-written message-passing systems (at least for simple tasks). That is, I would expect same program to run on say workstation cluster and CM-5. I know that's asking for too much but I am glad there are already some efforts on this direction. Short term: I Direct TCP/IP link between tasks. II PvmDataInPlace We should be thankful to Pvm_Team for giving us a good tool. Sanjay Kumar Graduate Student, Civil Engg. The Ohio State University E-mail : skumar@cad1.eng.ohio-state.edu ------------------------------------------------------------------------------- From: subba@lsil.com (Subba-Rao Kalari) > Are you using PVM today? We received PVM a couple of weeks back. We have not put it to use yet. > What are your application(s)? Our initial use will be understand the cost/benefit trade-off of distributing applications across the network. > How many and what kind of machines have you used with PVM? We will be using a heterogenous network of Sun and HP workstations. Thanks for making the results of your work available. Subba Rao ------------------------------------------------------------------------------- From: Elizabeth Yip Are you using PVM today? yes What are your application(s)? electromagnetics How many and what kind of machines have you used with PVM? iPSC/860 iPSC/2 Cray Sun Any comments and suggestions for improvement are welcome. How about an I/O package e.g. cread cwrite .... in the iPSC's ??? ------------------------------------------------------------------------------- From: csfb1!cowboys2!nstrauss@uunet.UU.NET (Neil Strauss) > > Are you using PVM today? We are currently evaluating PVM as an alternative to a proprietary tool` for parallelizing complex financial calculations related to Mortgage Backed Securities. There are several other tools under evaluation for the same purpose including Mentat, Linda, and various Symmetric MultiProcessor UNIX machines. > > What are your application(s)? see above > > How many and what kind of machines have you used with PVM? We have a network of several hundred SUN IPXes, and we have run parallel computations across 150 machines simultaneously using our proprietary coarse grained system. We have tested some calculations across a smaller network of 25 SUN IPCs and IPXs. ------------------------------------------------------------------------------- From: wiener@haze.rap.ucar.EDU (Gerry Wiener) Are you using PVM today? Not currently. What are your application(s)? Meteorological How many and what kind of machines have you used with PVM? None Thanks for your efforts. Gerry Wiener Email:gerry@ncar.ucar.edu ------------------------------------------------------------------------------- From: rwb@c3.crd.ge.com (Robert W. Benway) Are you using PVM today? *But-of-COURSE* What are your application(s)? Electromagnetics, fluid mechanics, medical diagnostic imaging, & others How many and what kind of machines have you used with PVM? 100-200 different systems. Suns, DECstations, HP-700s, Convex, CRAYs, SGI Any comments and suggestions for improvement are welcome. Comments (& brainstorms) have been given to ORNL/UTK as they occur.... -RWB ------------------------------------------------------------------------------- From: huangsl@stnick.me.vt.edu (100G Randolph) I am trying to use PVM today. My application is hydrodynamic instability ( it can be reduced to an algebraic eigenvalue problem ). I can access 9 RS6000. I just begin to use PVM, so I have no comment at this time. ------------------------------------------------------------------------------- From: ychung@fcusqnt.fcu.edu.tw (Dr. CHUNG Yeh-Ching) Yes, i am using PVM. Finite element computing. a cluster of workstations (4 - 16). Suggestion: 1. I think the communication primitives need to be enhanced, i.e., provides more communication functions. 2. It seems to me that everyone wants to run PVM must install PVM in their directories. Otherwise, you cannot execute pvm daemon. 3. The interface between users and pvm can be enhanced. Yeh-Ching ------------------------------------------------------------------------------- From: I am using pvm3.1 today I have 3 SUN4's, 1 HP-370 and 1 SGI/Indigo I am dividing up a finite difference algorithm, it must do a SEND/RECEIVE to each neighbor processor upon each entry into the main loop. I don't like the way the master processor will choose itself as a subprocess as a slave, this seems to defeat the purpose of pvm, I could have just caspawned a pro a process, or called a subroutine to do the same. I seem to have problems if i call a SEND to another processor which has not yet made it to RECEIVE, my program takes a long time to run. If I let the RECEIVE processor be the HP, which is much faster, then it works fine, since it is already to accept the SEND processors data. Is their a reason for this ? Are their ways to make each processor stay in sync with each other ? I am using the fortran commands, I only have a manual for 3.0, did 3.1 add many toe others ? ------------------------------------------------------------------------------- From: Armaghan Hoveydai > Are you using PVM today? > Yes > What are your application(s)? > Teaching parallel programming concepts to undergradute students. > How many and what kind of machines have you used with PVM? > 10-15 SUN SPARCstations Regards, +============================================================================+ | Armaghan Ross Hoveydai A.Hoveydai@cit.gu.edu.au| | Manager Computing Support Ph +61 7 875 5027 | | Faculty of Science and Technology Fax +61 7 875 5425 | | Griffith University, | | Nathan, QLD, 4111, | | Australia. | +============================================================================+ ------------------------------------------------------------------------------- From: Michael Rezny Yes I am using PVM today, in fact regularly. My applications are research into distributed processing and in particular modelling drought conditions in Australia for the Department of Primary Industries. We have used PVM on SUN4 and SGI, I have also used it on DEC and HPas well I have ported it for use with the MasPar and LINUX ( a unix clone for 386/486 IBM PCs) Mike REZNY ------------------------------------------------------------------------------- From: nhan@tiny.me.su.oz.au (Nhan Phan-Thien) 1. Yes, I'm current using PVM3.1 2. The application is three-dimensional Stokes flows past a large number of particles be an indirect boundary element method (called completed double layer boundary element integral method). The problem is large in scale (up to 200000 3D boundary elements). 3. I'm using PVM on a network of o up to 4 DEC alpha o two Titan Ardent o up to 6 SGI Indigo o up to 22 DEC 5000/200 The system works fine, but communication is sometimes a problem. We are currently write up a paper on the application of the method with PVM. Thanks you again for a nicely developed software. Nhan Phan-Thien, Professor of Mechanical Engineering. ------------------------------------------------------------------------------- From: glennw@glenrowan.acci.COM.AU (Glenn Wightwick) > Are you using PVM today? Yes, I am using PVM 2.4.2 and PVM 3.1 > > What are your application(s)? Applications are in meteorology and the environment and sorting. > > How many and what kind of machines have you used with PVM? I am using PVM with a cluster of 16 RISC System/6000 Model 350s and 1 RISC System/6000 Model 52H. ------------------------------------------------------------------------------- From: kinoshi@crayamid.cray.com (Toshihiro Kinoshita) > > Are you using PVM today? > Yes, I am. > What are your application(s)? > QCD (Quantum Chromodynamics) > How many and what kind of machines have you used with PVM? > CRAY Y-MP SGI IRIS 4D Regards, - Toshihiro Kinoshita (kinoshi@cray.com) ------------------------------------------------------------------------------- From: nakano@scs.ased.mt.nec.co.jp (Eiichi Nakano) >>Are you using PVM today? No, but I will try soon. >> >>What are your application(s)? Mathmatics library. >> >>How many and what kind of machines have you used with PVM? Three or four. Machine EWS4800,UP4800 (This machine is made by NEC cooporation. ------------------------------------------------------------------------------- From: obayashi@vdrl.src.mei.co.jp (Yoshimasa Obayashi) l Are you using PVM today? No. But I usually use PVM3.1 two days per week. l What are your application(s)? Mainly, Matrix Computations ( Eigenvalue Problem, Solving Linear Systems, FFT etc ). I am planning for using Fluid dynamics and FEM now. l How many and what kind of machines have you used with PVM? SUN4. l Any comments and suggestions for improvement are welcome. l Please send the information to pvm-survey@cs.utk.edu. l Your help is much appreciated. If I find the bugs or the errors, may I inform it to the same E-mail address ? Yoshimasa Obayashi VLSI Devices Research Laboratory Semiconductor Research Center Matsushita Electric Industrial Co., Ltd. E-mail:obayashi@vdrl.src.mei.co.jp ------------------------------------------------------------------------------- From: Sergey Ten Dear collegues, We are sending you the answers on your questions about our PVM interests: - yes, we are using PVM; - we are using PVM both in research and educational processes; - today PVM system is installed on the part of local network of University of Aizu. This part includes four SUN-4 SPARC stations. May be in future it will be possible to spread PVM system widely in this local network (it depends from and needs the special permission). P.S. Would you be so kind to inform us about the easiest way to obtain information about: 1. PVM-group (we have heard about its existance); 2. Have this group special pool (available through ftp, for instance) of PVM applications; 3. Does there exist PVM news-list and how can we obtain it. Thank you in advance, Sincerely yours, S.ergey Ten ------------------------------------------------------------------------------- From: Koichi Konishi |> Are you using PVM today? Not as of today. I used it for a few experimental applications. |> What are your application(s)? Work scheduler using genetic algorithm, trajectory calculation of mass points, ... |> How many and what kind of machines have you used with PVM? NEC EWS4800 workstations and SPARCstations. Regards, Koichi Konishi Computer System Lab., C&C Labs. Phone: +81-44-856-2178, FAX: +81-44-856-2231 ------------------------------------------------------------------------------- From: Brian J. Keefe >Are you using PVM today? We are attempting to centrally administer PVM and HeNCE on a network of 300+ workstation or x-terminals. I have prototyped several applications with it. Others have done more extensive number crunching. >What are your application(s)? I personally used it to develop threaded PVM nodes with Concurrent C and the C++ tasking library for prototype purposes. I tested lwp with RPC but could have and should have tried it with PVM. The goal was/is to maintain a single PVM which varied applications utilize as a network communications substrate, as an alternative to RPC or socket programming. PVM seems to potentially provide a happy medium between the RPC and sockets paradigms. >How many and what kind of machines have you used with PVM? Our network is currently homogeneous. I've prototyped my applications on 2 or 3 machines only. Others in the organization use PVM for production applications on, I believe, more machines. >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. - Allow a PVM to dynamically accept registrations from different users with some nominal security - Explicitly support threads within PVM processes - Allow PVM daemons to initiate joining an existing PVM using some well-defined PVM id of some kind, possibly using DNS services to in some way identify existing PVMs. - Support a PVMinetd to start remote PVM daemons -- Brian Keefe m1bjk00@fed.frb.gov ------------------------------------------------------------------------------- From: Yogesh S. Wagle > Are you using PVM today? Yeah, but on an experimental basis. > > What are your application(s)? Currnetly I am running the SIMPLE benchmark. I will be also running another fluid dynamics application soon. > > How many and what kind of machines have you used with PVM? 16. all SUN4s > > Any comments and suggestions for improvement are welcome. Improve the process-processor mapping algorithm. It might be possible to monitor the load on the machines before spawning the processes onto them. -- Yogesh Wagle ------------------------------------------------------------------------------- From: ricardo@ing.puc.cl (Ricardo Schmutzer von Oldershausen) > > Are you using PVM today? Yes. > > What are your application(s)? I am using PVM to parallize an optimization engine (SIMPLEX method) as my thesis. > > How many and what kind of machines have you used with PVM? DEC 5000s, a SparcStation 1, a SUN Workstation, a PC-386 (DTK) (up to 8 machines simultaneously). Ultrix 4.3.0, Ultrix 4.2.0, Ultrix 4.1.0, SunOS 4.1.2, Linux 0.99p9 > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. Thank you for making PVM, it has helped me a lot with my thesis ! I ported PVM to Linux (took the BSD Makefile and had to modify just a few lines in 2-3 files, so that it compiled correctly. I have used it until now only in a stand-alone machine (at home ;-), but will propably port it to Linux 0.99p12 in two-three weeks in a networked configuration (together with the above machines). If you are interested in the port, please let me know, to send you the corresponding patch. Wishes ? Perhaps, a function to join two send buffers, let me explain why: I would like to be able to modularize software written under PVM, for example for writing libraries for PVM. These libraries should be able to receive send buffers as arguments and be able to include it's own data in front of the "user" data. Let's suppose you have a library of functions that uses PVM (in my case a pool of workers). These functions receive in a send buffer the data that must be sent to another process, but the library itself must pass along additional data AT THE BEGINNING of the message, so that the library stub at the receiver side can decode it first. At the moment, I must pass the user data in a structure and a user function that packs the structure, so that the library can make a send buffer, pack it own data and then call the user function to pack the user data. This arrangement is ineficient, because an additional copy of the data is required (and it would be nicer if you could just pass a send buffer descriptor to PVM libraries). Sometimes inside the library I recognize that there are many small buffers (is there a way to check the size of a buffer BEFORE sending it ?) for the same destination. Here also, it would be nice to be able to concanate all the buffers to send them in just one message (i.e. to send it as one big TCP packet, instead of sending many small ones. I suppose the PVM library itself does not make such an optimization (yet? ;-). So the user should be able to join buffers (of course this could be made when the buffers are made, but in my case when I pack the buffers I do not know yet to whom I will send them (this is resolved in a lower level where the load of each host is known, i.e. inside the library). Something like: int pvm_pkbuf(int bufid) Packs into the active send buffer all data members from the buffer bufid. Eventually, if it necessary for efficiency, the buffer bufid may be destroyed in the process (as if pvm_freebuf would have been called). Perhaps, in this way it is just necessary to change some pointers to make the operation (I haven't seen your code in detail ! I may be talking nonesense !!). Thank you ! Ricardo -- +----------------------+---------------------------------+--------------+ | Ricardo A. Schmutzer | Internet: ricardo@ing.puc.cl | Conectividad | | PUC, Santiago, Chile | | Distr. Comp. | +----------------------+---------------------------------+--------------+ ------------------------------------------------------------------------------- From: jjacob@paxvax.EE.CORNELL.EDU (Joseph Jacob) >Are you using PVM today? > Yes. >What are your application(s)? > I am experimenting with a particle simulation as a case study for our dynamic load balancing scheme for distributed computers. >How many and what kind of machines have you used with PVM? I've been running my simulations on up to 16 Sun Sparc 1+ workstations (PVM_ARCH = SUN4). > >Any comments and suggestions for improvement are welcome. Notes: 1) I found that my application can adversely affect other machines on the same network. One of my colleagues has reported that his workstation hangs and he gets the message "Ethernet jammed" when I run my program. Others have reported that I have no effect on them, even when their machine is in my PVM configuration. This may be due to the relatively heavy communication required in my particle simulation, rather than some problem with PVM. 2) I have been unable to get PVM to work for the following: a) two or more communicating processes on the same Sparcstation. b) ANY configuration of HP9000 700 series workstations (PVM_ARCH = HPPA) 3) The PVM documentation I have states that I should get a message like "PVM ready" when PVM is started with "pvmd3 hosts" (hosts is the host file). Then only message I get is something like "Tid ready", where Tid is the task ID of the process spawned on the same machine as pvmd3. I then have to keep checking the pvm console until the full configuration is in place. Thank-you. -- Joe Jacob ------------------------------------------------------------------------------- From: Toyama Sakuji >>Are you using PVM today? No >>What are your application(s)? Fasta >>How many and what kind of machines have you used with PVM? 5 Sparcstations, Sunos 4.1.2 Toyama Sakuji Institute for Virus Research Email stoyama@virus.kyoto-u.ac.jp ------------------------------------------------------------------------------- From: dunkel@carbon.chem.utah.edu (Reinhard Dunkel) I used PVM 2.? during the development of my analysis software for 2D INADEQUATE NMR spectra. A typical run was 70 RS/6000 CPU hours and I used my access rights to the physics student lab (20 RS/6000) to finish a run over night. I encountered numerous problems during with PVM 2.? (1) Starting pvmd required that all enumerated computers were really accessible. Pvmd did not recover from an unavailable machine and so I had to try several times (5 minutes each) to start pvmd until all currently "bad" computers were removed and pvmd would start. (2) Using a few local computers worked fine with PVM. When using a larger number of computers (>10) or remote computers I always ran into network problems. These problems were not caused by the average load that my application produced. Anyhow, PVM complained so much that it was impossible to determine whether or not any real progress was made. (3) Permission problems. I never figured out under what conditions computers could be used with pvmd. I am talking exclusively about RS/6000s here. To specify passwords from the .netrc file worked more reliably than to depend on any rsh permissions - so that's what I used. But lots of RS/6000s would not cooperate. I suspect the "secure network" option used on some RS/6000s was the problem but I never really figured it out. I did not find a section in the PVM documentation that explained the communication of the pvmd daemons and how the computers have to be administered to allow it to work. (4) In PVM 2.? I did not find a smooth way to tell from my program whether pvmd was running or not. Enrolling in PVM should fail with a "problem" return code if pmvd is not running rather than for PVM to print lots of error message and to die. So I had to modify the source code substantially (don't print error messages and time out quickly) to make it work. (5) When PVM 3.0 came along (February) lots of things which worked in 2.? did not in 3.0 and lots of stuff described in the new manual did not work either. I reported all problems I found and Robert Manchek (manchek@cs.utk.edu) confirmed that these were indeed problems. I didn't try a newer PVM version ever since. I improved the speed of my software so one run can be finished over night on one RS/6000. I found the over-night-run to be more practical than to use PVM. So PVM is supported in our software but nobody uses it regularly as far as I can tell. As datasets get bigger and fancier approaches to the analysis are introduced the run times of the analysis go up. If analysis times go over 10 CPU hours again I am sure PVM will get more attractive and I hope I'll discover that PVM got less painful to handle in the newer versions available now. __ _ | ) o | | | \ | | | / _ |_ _ _ _| | | | _ | |/\ /_) | ^ ^ | | _) | / | | | | | ^ ^ |_) /_) | | \_/\___|_| | |_| |_(_|__|___\_| |_/__|_|_| | |_| \_/\___| Department of Chemistry Office: (801) 581-7351 2020 Henry Eyring Building b113 Home: (801) 582-7516 University of Utah FAX: (801) 581-8433 Salt Lake City, UT 84112 dunkel@chemistry.chem.utah.edu ------------------------------------------------------------------------------- From: bessey@milo.math.scarolina.edu (Al Bessey) Are you using PVM today? Yes What are your application(s)? I am using pvm to write a parallel version of a Lagrangian Particle Dispersion Model which is a part of a visualization package for the Regional Atmospheric Modelling System (RAMS) code. I intend to also use pvm for rewriting and parallelizing the RAMS code as part of my PhD thesis work. We at the University of South Carolina have an Intel Paragon A4 and eagerly await the pvm port for that machine so that we can initially port our pvm codes with few if any modifications to our 56 node machine and then to the 512+ node machine at Oak Ridge. Other people here have been using pvm on other atmospheric and ground water models as well and also await the port. How many and what kind of machines have you used with PVM? We mainly use a subnet of 7 IBM RS/6000 workstations but are getting a new bunch of SGI equipment that may be used, mostly for image processing. Also, as I have said, we want to use it extensively on the 56 and 512+ node paragons. Any guess as to when that port will be available? Any comments and suggestions for improvement are welcome. I have a few questions on related topics. I was considering writing a third party PVM tutorial type book. I have had to give introductory type mini courses and I am trying to keep all notes and examples I use for part of this endeavor. What permissions or approvals do I need from your team? I may also need some assistance in proofing and in critiquing and was wondering if we could work something out for this as well? We also are still heavily using pvm 2.4 mainly because of the name changes in 3.x versions. Do you suspect any more such renamings in later upgrades? Also, how long will the pvm software be supported as it is now? I think pvm is a great package and thanks in advance for you help with all my questions. Al Bessey bessey@milo.math.scarolina.edu (803) 777 - 9333 ------------------------------------------------------------------------------- From: arita@mrit.mei.co.jp >Are you using PVM today? Yes. >What are your application(s)? Ising Model. >How many and what kind of machines have you used with PVM? Next , SparcStation2 , SparcStation10 , SolbourneSeries5 arita@mrit.mei.co.jp Panasonic/Japan ------------------------------------------------------------------------------- From: frank@photon.ansto.gov.au (Frank Crawford) > > Are you using PVM today? Yes and no. I've installed it tested it (PVM 3.1) and am waiting on a colleague to start a project using it. > What are your application(s)? The planned applications are all demonstration projects. Primarily comparison of vectorised performance with parallel performance (on both a CM-5 and under PVM). The initial project will be for a diffusion code. > How many and what kind of machines have you used with PVM? Primarily on SGI's and Sun's, however I have previously ported PVM 2.4 to Fujitsu VP2200, i486's running SVR4 and a 68020 based SVR3 system. (I'll port PVM 3.1 to these in the near future, when the colleague mentioned above finds some time to use them.) > Any comments and suggestions for improvement are welcome. Making porting easier. Particularly, the entries in `archcode.c'. While the entries themselves are easy, patching the file for each change on each host gets to be a pain. Some method of updating the types that doesn't need to be compiled in would be nice. > Your help is much appreciated. Your welcome. Frank Crawford ------------------------------------------------------------------------------- From: "Dano Kral" > > Are you using PVM today? We are not using PVM now. > > What are your application(s)? There are calculations in quantum chemistry. > > How many and what kind of machines have you used with PVM? We have only convex C3400 now, but on end of year we will have two HP 735 procesors and we wont create meta cluster C3400 + 2 HP. Regards, Dano. ------------------------------------------------------------------------------- From: j011@cray.op.dlr.de (Leisen Siegfried) 1. I'm not using PVM today in production 2. This is a German research lab and the people are thinking about distributed computing environments in general. They have a few workstations in the computing centre and a lot of Suns and RS6000s in the different institutes. 3. Our first test of PVM was running a mandelbrot set on a CRAY Y-MP and a RS6000/580 and a RS6000/550. I'm a software site analyst from CRAY Research and the people at DLR are playing around with PVM. I did only the installation on the CRAY system. Regards Siegfried Leisen Siegfried Leisen, CRAY Research GmbH e-mail: j011@sn1414.op.dlr.de c/o: DLR Rechenzentrum lei@crmunich0.cray.com Muenchner Str. 20 j011@vm.op.dlr.de 82234 Wessling phone: +49 8153 2032 Germany ------------------------------------------------------------------------------- From: worsch@ira.uka.de pvm> Are you using PVM today? No. pvm> What are your application(s)? We used pvm for moving data from a process running from a machine with dedicated hardware to a process running on another machine. pvm> How many and what kind of machines have you used with PVM? We used pvm only on Sun4 machines. pvm> Any comments and suggestions for improvement are welcome. We would have welcome better documentation (e.g. on which library routine uses which data transportation mechanism). pvm> Your help is much appreciated. You're welcome. Have a nice day. Thomas Worsch ------------------------------------------------------------------------------- From: awidmann@ifp.mat.ethz.ch > Are you using PVM today? Not today but I was using it several weeks to become familiar with it. Unfortunately we still don't have NFS installed why it is a little bit troublesome to work with PVM. As soon as we will have installed the network services NIS and NFS, I will do productive work with PVM. > What are your application(s)? Prediction of properties of polymers by computer simulation. Some of the algorithms we use (developed at our institute) are very suitable for parallelization (e. g. on Alliant or Sequent Symmetry systems). > How many and what kind of machines have you used with PVM? I have been using PVM on our 9 SiliconGraphics workstations (Crimsons and Indigos). As soon as NFS works fine, I will also try to use it on IBM AIX/6000 systems. Regards ------------------------------------------------------------------------- Albert H. Widmann Institut fuer Polymere "The best way ETH Zentrum, CNB E98.2 to predict the future CH - 8092 Zuerich is to invent it." Tel: +41 1 256 5672 FAX: +41 1 261 0215 Alan Kay e-mail: awidmann@IfP.mat.ethz.ch ------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: yaniv@bilbo.fiz.huji.ac.il (Yaniv Romem) I am not using PVM on a regular basis, I have used it in the past few months a few times in order to construct and test parallelizations of programs which were afterwards run on machines like the CM-5. The kinds of applications are generally physics programs. I have used configurations of DEC's (MIPS) , HP's (720) and SUNS (various) though I usually ran on 1-3 computers at a time. By the way I am working on an implementation of PVM on a local parallel computer we have here at the Hebrew University (called the MAKBILAN). I am practically done, although I had to cut out some of the more exotic functions like signals because of the architecture of the machine. Best wishes, Yaniv Romem ------------------------------------------------------------------------------- From: lapique@siisg1.epfl.ch (Lapique) > > Are you using PVM today? Yes > > What are your application(s)? > LU Decomposition, matrix product > How many and what kind of machines have you used with PVM? > 20 : Sun /FDDI, Silicon Graphics/Ethernet Cray-2 > Any comments and suggestions for improvement are welcome. > improve FDDI integration. -- Francis Lapique-EPFL/Computer Center ----------------------------------------------------------------- Swiss Federal Institute of Technology-Lausanne E-Mail : lapique@sic.epfl.ch/lapique@siisg1.epfl.ch Voice-Mail : (41)-21-693-45-96 Fax: (41)-21-693-22-20 ----------------------------------------------------------------- ------------------------------------------------------------------------------- From: fujita@hip.atr.co.jp >> Are you using PVM today? No. >> What are your application(s)? Only example, now. But maybe speech recognition. >> How many and what kind of machines have you used with PVM? 10 Sun's. >> Any comments and suggestions for improvement are welcome. I am studying about PVM and parallel computing. - Fujita ATR Human Information Processing Research Labs. ------------------------------------------------------------------------------- From: fadb@corona.oslo.dnmi.no (Dag Bjoerge) Yes, I'm using PVM. Until now I have only run some small tests. We are going to implement the Nordic+ weather forecasting system HIRLAM in PVM. I have just installed PVM on my SGI, and done some testing. I have done some programming on an Intel Paragon using their 'nx' programming system, but PVM on the Paragon has given me some problems that I have not yet solved. I am a to novice PVM user to give you any valueable suggestions, but I send you my best wishes for your project. Dag Bjoerge The Norwegian Meteorological Institute fadb@corona.oslo.dnmi.no ------------------------------------------------------------------------------- From: "Craig A. Lee" > >Are you using PVM today? Yes. >What are your application(s)? (1) Parallel discrete event simulation (2) Cloud depiction and forecasting, (via CC++). >How many and what kind of machines have you used with PVM? LAN of suns and SGIs (iris), CM-5, KSR, and the iPSC/860 and Touchstone Delta (when I get around to finishing the config of g++) >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. I eventually will need to have asynchronous receive (ala the Intel hrecv) for some algorithms. I know that mcavende@ringer.cs.utsa.edu (Mark Cavender) has worked on this for some machines/OSs. Your welcome, --Craig ------------------------------------------------------------------------------- From: Peter Junglas => Are you using PVM today? => yes, I'm just preparing a talk on the Convex Cluster Software (for Convex and HP machines), which includes PVM. => What are your application(s)? => Since I'm working at a computer centre, I have no applications myself, but use PVM for teaching. People here at the TU use PVM for several engeneering applications such as finite element methods, simulation of semiconductors and image processing. I will forward your letter to them. => How many and what kind of machines have you used with PVM? => At the computer centre PVM is installed on a Convex C3840, a Convex C120, a Silicon Pool (3x Indigo and 8x 4d) and a HP 9000/700 Cluster (6x 730, 3 smaller machines). => Any comments and suggestions for improvement are welcome. We need some tools profiling and analyzing tools. When will Xab be ported to PVM3 ? A tool like Paragraph - or a better conversion to PICL format - would be helpful. Greetings from Hamburg Peter -- Peter Junglas Tel.: 040/7718-3193 TU Hamburg-Harburg, Fax.: 040/7718-2803 Rechenzentrum email: Junglas@tu-harburg.d400.de Denickestr. 17 21071 Hamburg ------------------------------------------------------------------------------- From: heinrich@vnet.IBM.COM To: pvm-survey@cs.utk.edu I'm using pvm for self-studying parallel algorithms and things like that. I use for about a year, on and off, depending on other task I've to do. I use IBM RISC System/6000 (for self-study 1 or 2), tried it with 4 .. 8 in a dedicated environment (IBM Rome). Regards, Dr. Heinrich Backhausen IBM Deutschland Informationssysteme GmbH MZ Duesseldorf Germany ------------------------------------------------------------------------------- From: K. Visser > Are you using PVM today? Not at the moment. It takes some time to get acquanted with the software and the people at our department who should use it, simple didn't find the time for it yet. > What are your application(s)? We want to use pvm together with xadaptor to create the sourcefiles more easily. We haven't tried this interface yet. > How many and what kind of machines have you used with PVM? 4xHP9000/720, works fine! Greetings, Kees Visser -- University of Groningen Department of Mathematics P.O. Box 800 9700 AV Groningen The Netherlands Telephone: +31 50 633987 (secretary) +31 50 633968 (extension) +31 50 633976 (telefax) Email: K.Visser@math.rug.nl ------------------------------------------------------------------------------- From: roberto@csv.ica.uni-stuttgart.de (Roberto Grosso) > Are you using PVM today? We carry out some times a presentation, where interprocess communication is based on pvm. > > What are your application(s)? The presentation is a distributed supercomputer CFD simulation and visualization, where many parties are able to interact with the application: Computer Supported Cooperative Work. One presentation was distributed over around 500 Km with VBN, FDDI and ATM network technologies. > > How many and what kind of machines have you used with PVM? > The application includes a Cray-2 as a number crunching, a Cray YMP as a file server, and many SGI workstations for the visualization. > Any comments and suggestions for improvement are welcome. The application uses pvm version 2, so I don't know which is the actual status of pvm. If one has many processes in one machine which communicate with some other processes in other machine, the bottle neck in the whole application is the pvm daemon! Sincerely yours roberto -- Roberto Grosso grosso@rus.uni-stuttgart.de ICA II - University of Stuttgart Tel.: +49 (0) 711 / 685 5996 Pfaffenwaldring 27 Fax : +49 (0) 711 / 678 7626 70550 Stuttgart, Germany ------------------------------------------------------------------------------- From: zrw@rz.uni-jena.de (Renate Wahl) Hello, dear colleagues, because I've hollidays soon I'll answer at once. I'm working in the computer center of the university in Jena, Thuringia, in East Germany. > Are you using PVM today? > > What are your application(s)? I've installed PVM and today I'm testing the system and preparing a course for our users. Specially I think of users from Physics and Chemistry with large and cpu-intensive programms with different requests. And also I think of students. I believe, in october we make it public. > > How many and what kind of machines have you used with PVM? > In "my" virtual pvm - machine are the following machines: 1 x IBM RISC 6000/560, 5 x RISC 6000/22W, 1 x RISC 6000/340, 1 x RISC 6000/320 1 x DEC 5100, 1 x DEC 3000 AXP 500, 1 x Silicon Graphics - workstation Best wishes to the PVM Project Team Renate Wahl -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ I E - MAIL : zrw@hpux.rz.uni-jena.de I Renate Wahl I I Phone : ( +49 03641) 636290 I Rechenzentrum der I I FAX : ( +49 03641) 22435 I FSU Jena I I I Humboldtstr. 2 I I I 07743 Jena I I I Germany I -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ------------------------------------------------------------------------------- From: stoessel@irsun21.ifp.fr (Alain Stoessel) Hi PVM Project Team, First thanks for your nice job on PVM3. We, at Institut Francais du Petrole, used PVM for 18 months. At the beginning, we used it to program a network of workstations. Because of the performance of the network and problems of load balancing, we don't use this paradigm for real calculations. (We have also a supercomputer Fujitsu VP2400 so it is difficult to go faster on a cluster). Because of the kind of applications we work on (seismic simulation, reservoir modelling, CFD and DNS of combustion), we now prepare the migration of these codes to parallel architecture. We have acces to remote parallel ressources (CM-5, AP-1000,IPSC/860..) but we still develop our algorithms on our workstations and on Fujitsu with PVM. Moreover, in order to have portability, we write our codes by using 2 or 3 libraries for message passing.( generally PVM, Parmacs and Casim (AP1000)). When MPI will be available........... We use PVM in other ways: - First, to communicate between workstations and Supercomputer in order to make visualization.... - To make 'out-core' calculations on our Fujitsu. We use PVM to make asynchronous I/O for managing the different levels of storage. At IFP, PVM is running on: - SUN workstations - SGI workstations - Fujitsu VP2400 supercomputer (in-house porting: lots of problems with BSD packages immediate since 3.1.5 and SUN4SOL2 availability) (I can communicate Makefiles) Comments: --------- We like PVM but we have some suggestions: - quick availability of DataInPlace buffers (no need to build buffers..) - integration of a tool for profiling and debugging (like Adam's tool XAB for fortran programs). Compatibility with Paragraph's tracefile. Bye. +-----------------------+------------------------------+ | Alain STOESSEL | Institut Francais du Petrole | | Tel: 33.1.47.52.71.33 | Parallel processing group | | Fax: 33.1.47.52.70.22 | 1-4 Av de Bois-Preau | | | 92506 RUEIL-MALMAISON | +-----------------------+------------------------------+ | Email: stoessel@irsun21.ifp.fr | +-----------------------+------------------------------+ ------------------------------------------------------------------------------- From: Ulrich Mehlhaus Your pvm survey We use pvm (3.1.4) for distributed simulation of robotic tasks. Therefor we created different processes responsible for instance to emulate behaviour of manipulators, effectors, vehicles, etc. . All these processes communicate via pvm, where all the processes are instantiated by seperate unix cammands (i.e., they are not spawned by a central master). To resolve the target tid, we've developed an extension of your group server which could be offered to the public if interested (see below for a description). Currently we use Sun Sparc (OS 4.1.2) and Silicon Graphics for our applications, maybe we are going to port the pvm system to Evans & Sutherland. Let me add a point of critics at the end: quite a while ago we've detected a bug in pvm_probe (and reported to Adam, who confirmed and promised to correct it), but even in the current release it doesn't seem to work the way it is described in the manual. So maybe in the NEAR future .... ? Best Regards Uli ------------------------------------------------------------------------------ Ulrich Mehlhaus Institut fuer Prozessrechentechnik & Robotik Universitaet Karlsruhe Kaiserstrasse 12 Postfach 6980 76128 Karlsruhe email: mehlhaus@ira.uka.de tel: 0721/6084243 fax : 0721/606740 Dear PVM users, we've just upgraded our specialized PVM Group Server to PVM 3.1.3 and we think that maybe some of you could be interested in it. Purpose of our Group Server: assume that we have two processes wanting to exchange messages. Both are started from the shell or from other programs, but they have NO relation to each other (i.e. not spawned or forked). The only things they know from each other: 1. the other one is running as a PVM process 2. and if alive, the name of the other process. Using pure PVM without Group Server is not able to resolve the tid of the other process, so we can't exchange messages. Using the Group Server coming with the PVM package offers one possibility to retrieve the desired information: /* process 1*/ /*process 2*/ mytid = pvm_mytid ... mytid = pvm_mytid ... pvm_joingroup("process1"); pvm_joingroup("process2"); pvm_joingroup("everyone"); pvm_joingroup("everyone"); pvm_barrier("everyone",2); pvm_barrier("everyone",2); /* both are ready and running */ othertid = pvm_gettid("process2"); . . . . . . . This leads to a set of one-elemented groups, which only serve for tid resolving. In our approach we changed and added some of the group functions allowing us to store and to retrieve also process names/filenames: a. Change function "pvm_joingroup(group)" to "pvm_joingroup(group,fn)". Every process must specify his own filename and the group he wants to join. b. Add the function "int pvm_gettid_fn(group, fn)". Every process can call the function to get the TID of a file in a group. c. Add the function "int pvm_gettid_fn_inst(group, fn, inst)". Every process can call the function to get the TID of a file with instance no. inst in a group. Our small changes to the original Group Server lead to a more comfortable usage of the group simply by using some new group functions. This Group Server is only useful in the area of independently instantiated processes, so if you spawn your PVM processes, forget about. We would like to offer the sources per e-mail request. The code is provided ``as is'' without express or implied warranty. In addition, we can't guaranty to upgrade the software according to PVM upgrades. The software was tested only on SUN3 and SUN4, but porting it to other architectures should be easy. If interested, just drop me an e-mail to mehlhaus@ira.uka.de Best Regards Uli ------------------------------------------------------------------------------ Ulrich Mehlhaus Institut fuer Prozessrechentechnik & Robotik Universitaet Karlsruhe Kaiserstrasse 12 Postfach 6980 76128 Karlsruhe email: mehlhaus@ira.uka.de tel: 0721/6084243 fax : 0721/606740 ------------------------------------------------------------------------------- From: Georg Stellner > Are you using PVM today? Yes. > > What are your application(s)? No concrete application. I am working on tools and messaage-passing environments for coupled workstations. Consequently I only have small applications for testing purposes. > > How many and what kind of machines have you used with PVM? 35 Sun Sparc, 110 HP Model 715 -- *** Georg Stellner stellner@informatik.tu-muenchen.de *** *** Institut fuer Informatik, SAB phone: +49-89-2105-2689 *** *** Technische Universitaet Muenchen fax: +49-89-2105-8232 *** *** 80290 Muenchen room: S 1211 *** !!! ^^^^^^^^^^^^^^ new address, please update Your archives !!! ------------------------------------------------------------------------------- From: theo@macsch.com > Are you using PVM today? > no > What are your application(s)? > developing s/w for "Explicit Transient Dynamics" (F.E.): MSC/DYTRAN. This is a 100% local effort. > How many and what kind of machines have you used with PVM? > none so far. For ETD it is tough to do on a loosely coupled cluster, because there is communication every time step (~100 msec.). > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Right now we prefer solutions that are integrated within the compiler language. For minimum overhead, PVM calls could eventually be dealt with by the compiler and scheduled in the code. Any de-facto standard will do for us, as long as all manufacturers, including workstation manufacturers, support it. Important thing for us is STANDARDS. I could be wrong, but on the long run I see better future in HPF, with PVM as an intermediate, available today, and... something that works today. Perhaps we'll see PVM implemented within HPF; I don't know. If I had to make a choice today, I would probably pick PVM, since it is available on all platforms. best regards, Theo Groen, the MacNeal-Schwendler Corporation BV the Netherlands ------------------------------------------------------------------------------- From: pfschil@pasun1.GKSS.DE (Helmut Schiller) Are you using PVM today? yes What are your application(s)? Monte-Carlo ray tracing --> radiation transfer through cirrus clouds How many and what kind of machines have you used with PVM? about 40 SUN's, rarely RS6000 Thanks for your efforts! H. Schiller ------------------------------------------------------------------------------- From: Resch@rus.uni-stuttgart.de (Michael Resch) > > Are you using PVM today? yes > What are your application(s)? fluid mechanics > > How many and what kind of machines have you used with PVM? a cluster of 6 RS6K a PGON Michael Resch -- ------------------------------------------------------------------------------- Michael Resch Phone: ++49 (0)711-685-5834 Rechenzentrum Universitaet Stuttgart Fax: ++49 (0)711-678-8363 Numerical Methods for Supercomputers Allmandring 30 email: resch@rus.uni-stuttgart.de 70550 Stuttgart Germany ------------------------------------------------------------------------------- From: "Gerd Meister" > What are your application(s)? > Our group is working in the field of parallel discrete event simulation and esp. in parallel logic simulation. > Are you using PVM today? > We are only running a test installation of PVM but I am quite interested in evaluating our applications on top of PVM because our simulators are running (with the same basic code) on a network of SUN workstations and on the Intel Hypercube IPSC860. The currently used base for concurrent programming is a system from the Technical University of Munich called MMK that supports both architectures and is a small message passing kernel. Nevertheless, I start PVM in an iconified window as default application within my xsessions. > How many and what kind of machines have you used with PVM? > We are using 8 Sun4 Workstations (Sparc 2 and ELCs), one Sun 3/140, 2 Solbourne and an Intel Hypercube IPSC860 with 32 nodes. Kind regards, Gerd. -- Gerd Meister Tel.: 0681/302-4576 Universitaet des Saarlandes FB Informatik Fax : 0681/302-4421 AG Mattern (Geb. 36) Postfach 11 50 D-66041 Saarbruecken Email : meister@cs.uni-sb.de ------------------------------------------------------------------------------- From: Christoph Pleier > > Are you using PVM today? No. > What are your application(s)? Parallel Ray Tracing > > How many and what kind of machines have you used with PVM? Sun SPARCstations, HP 9000/7xx ----------------------------------------------------------------------------- | Christoph Pleier | Phone: (+49) 89-2105-8136 | | Institute of Computer Science | Fax: (+49) 89-2105-8180 | | Technical University of Munich | Email: pleierc@informatik.tu-muenchen.de | ----------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Volker Strell > Are you using PVM today? Yes > What are your application(s)? For my thesis I am implementing some parallel algorithms for Markov chain analysis. > How many and what kind of machines have you used with PVM? I am using up to 10 Sun SPARC Stations. Volker --- Volker Strell strell@christa.informatik.uni-dortmund.de ------------------------------------------------------------------------------- From: carsten@mistral.ifi.uni-klu.ac.at Carsten Weich Institut fuer Informatik, Universitaet Klagenfurt Austria Are you using PVM today? no What are your application(s)? we are using it as a base for higher-level parallel language constructs and libraries How many and what kind of machines have you used with PVM? currently we have 4 MIPS (DEC 5000/133) but within 2 months we will get an DEC/ALPHA-Farm consisting of 12 stations. I would be insterested in the result of this survey. Greetings from Austria! ------------------------------------------------------------------------------- From: jaho@dutlfs1.tudelft.nl (ir. J.M.A.M.Hol) > Are you using PVM today? Yes and no. Yes, for testing and development. Mainly to get a feeling for what PVM does and can do for us. No, not yet for production type of work. > What are your application(s)? My research group works in the field of structural mechanics with non-linear collapse load calculation of thin-walled shell structures as major topic. Leading researcher is prof. dr. Johann Arbocz. You can find publications of his work and work of his co-researchers in publications dating back to the late 60s and early 70s. At this moment I'm the one responsible for development of the DISDECO project (= Delft Interactive Shell DEsign COde). Basically this is an hierarchical design and analysis framework in which analysis modules ranging from simplified solutions of the applicable PDEs upto full-featured FD and FE packages are integrated in a transparent manner. Added to that is a modern interactive environment supported by database technology, graphics, on-line context sensitive help systems and in the future expert system technology. The DISDECO system runs on workstations and automagically executes analysis modules and programs either locally and on (appropriate) remote machines. One of the primary requirements is portablility. At this moment Sun SPARCs are the baseline systems with Sun SPARCservers and a Convex C-3840 as remote compute servers. We also have at our disposal an IBM mainframe as well as access to both Cray YMP and NEC SX-3 supercomputers. We use the DISDECO system locally for both research and educational purposes. Among the other sites using the system is NASA Langley. In that context I am constantly looking for, evaluating and testing software and hardware. My conclusion at this point is that I want to use PVM for both parallel computational work but also for master-slave type application work. We are going to develop and implement a pilot master-slave application in the context of the DISDECO project. The master being the main DISDECO processor running on Sun SPARC-machines while the slave will be a (remote) data-manager for geometric imperfection survey data running on a HP9000 system. At a later date we'll start experimenting with PVM for parallelizing our analysis programs. The reason for postponing this activity is a (much too) heavily loaded network which already is a bottleneck for our normal work. Since the local network is in the process of being redefined and upgraded I expect a much improved environment in the near feature which will allow efficient loosely coupled parallel computation. > How many and what kind of machines have you used with PVM? I have tested PVM (2.4.x, 3.0 & 3.1) primarily on a cluster of Sun SPARC-systems with Convex C-3840, Silicon Graphics and HP9000 systems as secondary targets. Both Sun and Convex appear to be working alright although intermachine connectivity is a problem. The (single) SGI-machine was only used for primary testing. The HP9000 is an older system running HPUX 7.x on which PVM will not install itself automagically. Since I need it to run on that specific machine one of my people will be looking into that later this year. > Any comments and suggestions for improvement are welcome. At this point we have too little practical experience to be able give sensible comments and/or suggestions. If something comes up we will certainly contact the PVM group. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. Your welcome although you may find my answers somewhat to verbose. With regards, +------- Jan Hol --------------------------- ir. J. M. A. M. Hol -------------+ |internet: jaho@dutlfs1.TUDelft.NL Structures Group | |bitnet: vlsthol@hdetud2.TUDelft.NL (backup) Faculty of Aerospace Engineering | |phone : +31-15-785379 ### # # #### # Delft University of Technology | |fax : +31-15-781822 # # # # # # P.O. Box 5058, 2600 GB Delft | |telex : 38151 butud nl # # #### # # # # The Netherlands | +----------------------- #### # # #### #### ---------------------------------+ ------------------------------------------------------------------------------- From: Ron Levkovitz Hi, We use PVM in the maths department of Brunel University, London, U.K. The major project is a parallel branch and bound algorithm for integer programming problems, the software is currently in development but is operational. Other projects include direct and iterative sparse systems solvers. We use PVM only on our SUN Sparc workstations. I hope this answers your questions. Ron Levkovitz ron.levkovitz@brunel.ac.uk ------------------------------------------------------------------------------- From: Xavier-Francois.Vigouroux@lip.ens-lyon.fr (Xavier-Francois Vigouroux) | |Are you using PVM today? YES, almost every day |What are your application(s)? parallel visualization of trace files |How many and what kind of machines have you used with PVM? 1, soon 2. |Any comments and suggestions for improvement are welcome. |Please send the information to pvm-survey@cs.utk.edu. |Your help is much appreciated. The way to install PVM 3.1 on iPSC is not clear at all. Could you make clearer explenation. Whatever, the soft is great. I congratulate your team. ------------------------------------------------------------------------------- From: pierre@sgirenne.rennes.sgi.com (pierre VERCRUYSSE) > > Are you using PVM today? not me but one customer IRCN at Nantes > > What are your application(s)? PERMAS, They want to do some test for paralelization with shared memory and distributed memory > > How many and what kind of machines have you used with PVM? only SGI INDIGO R3k and R4k ------------------------------------------------------------------------------- From: Wilhelms@Uni-Augsburg.DE (Gerhard Wilhelms) Dear PVM Project Team, At Augsburg University there is a small "Parallel and Distributed Computing Group" with one PhD student (me) and eight graduate students who all use PVM (daily) and HeNCE (almost every day). We are working on enhancements that make PVM easier to learn and use and on load balancing schemes with low overhead. We try to veryfy our results with applications from the theoretical Physics department. Currently we are using 10 IBM RS/6000 and 5 HP Series 400 (68040 based) computers. This number can (and will) be increased to 50+ RS/6000, 5 HP/400 and 5 HP/700 as our dynamic load balancing proves to be effective ... Hope that helps. Sincerely, Gerhard Wilhelms Augsburg University Department of Mathematics ------------------------------------------------------------------------------- From: Peter Dauphin > > Are you using PVM today? Yes, but not very extensively > > What are your application(s)? ray tracing algorithms > > How many and what kind of machines have you used with PVM? > 7 HP9000/700 about 10 Sun Sparc with SunOS 4.3.1 Best Wishes Peter Dauphin -- +--------------------------------------------------------------------------+ | Peter Dauphin Phone: nat. 09131/85-7026 | | Institute of Computer Science int. +49-9131/85-7026 | | University of Erlangen-Nuernberg Fax: +49-9131/857409 | | Martensstr. 3, D-91058 Erlangen, Germany | | email: pdauphin@immd7.informatik.uni-erlangen.de | | /C=de/A=dbp/P=uni-erlangen/OU=informatik/S=pdauphin | +--------------------------------------------------------------------------+ ------------------------------------------------------------------------------- From: ph@mema.ucl.ac.be (Peter Henriksen) Are you using PVM today? Yes.(2.4 and 3.1) What are your application(s)? Flow simulations using the Finite Element Method. How many and what kind of machines have you used with PVM? SG Iris/SG Indigo, HP 710/720/730/735, Intel Paragon, DG 200/6020, Sun 3/SPARC1/SPARC2/, CONVEX 240/3820, IBM RS6000 320, DEC 3100/5000. ------------------------------------------------------------------------------- From: D.Lamptey@sheffield.ac.uk ]Dear Colleague, ]Could you take a moment to answer three questions. ] Sure! ]Are you using PVM today? ] Yes. ]What are your application(s)? ] Large (coarse, grained) geological simulation (intended. see below) ]How many and what kind of machines have you used with PVM? ] I have used Sparc (Sun4) machines well with pvm for harnesses. I am having problems, though, getting the code to run under Solaris2. I am hopeful, though... ]Any comments and suggestions for improvement are welcome. I could not "make" pvm under Solaris, so I made it under sunos4.1x, and am trying to run it on solaris machines!! Keep it coming! DErryck. p.s. Can provide more info, if required.. _____________________________________________________________________ National Transputer Support Centre, Derryck Lamptey, MSc(Eng) (University of Sheffield) (D.Lamptey@sheffield.ac.uk) 5 Palmerston Road, Sheffield S10 2TE South Yorkshire, U.K. Telephone: (+441-742) 768740 (0742) 768740 (NTSC@sheffield.ac.uk) Facsimile: (+441/0)742 727563 "use e-mail: save a tree" _____________________________________________________________________ ------------------------------------------------------------------------------- From: demmer@prl.philips.nl > > Are you using PVM today? Yes > > What are your application(s)? "MOLECULAR DYNAMICS FOR SILICON" > > How many and what kind of machines have you used with PVM? 13 HP-UX A.09.01 E 9000/735 4 HP-UX A.09.01 E 9000/755 1 IBM/AIX RS/6000 -- e-mail: demmer@prl.philips.nl phone: +31 40 743080 mail : Stephan Demmer, Computer Services, Central Systems, Philips Research Laboratories, Prof.Holstlaan 4, (postbox WY 01), 5656 AA Eindhoven, The Netherlands. ------------------------------------------------------------------------------- From: thevenin@thor.iwr.uni-heidelberg.de (Dominique Thevenin) Answers to the survey : 1) yes, I am using pvm at present time. This constitutes about half of my job, as I am also working on ''classical'' parallel computers using native message-passing. 2) My domain is reactive Computational Fluid Dynamics (pollutant formation in flames, instability...). 3) We have at our lab a cluster of 5 Iris Silicon Graphics under Ethernet. Another neigbour institute has 6 RISC 6000 with optical link. I am using both at present time. Dominique *************************************************** * Dr Dominique Thevenin * *-------------------------------------------------* * Interdisziplinaeres Zentrum fuer * * wissenschaftliches Rechnen - IWR - * *-------------------------------------------------* * Universitaet Heidelberg * * Im Neuenheimer Feld 368 * * D-69120 Heidelberg, GERMANY * *-------------------------------------------------* * Tel. : (49) 6221/564984 * * Fax. : (49) 6221/565224 * * e-mail : thevenin@thor.iwr.uni-heidelberg.de * *************************************************** ------------------------------------------------------------------------------- From: fred@sobel.u-strasbg.fr (Frederic PIERRE) > Dear Colleague, Copy/paste :-) > > Are you using PVM today? > I'm still in the evaluation phase of a project where PVM could be used. > What are your application(s)? Parallel Discrete Event Simulation in the field of digital logic circuits. The main goal is to speed up simulations described in a high level langua ge such as VHDL. > > How many and what kind of machines have you used with PVM? Around 15 workstations, mainly Sparc Sun and IBM RS/6000. Best regards and many thanks for the work done. Fred. =========================================================================== Frederic PIERRE. ENSPS/LSIT 7 rue de l'universite F-67000 Strasbourg FRANCE Tel: (33) 88 35 80 84 Fax: (33) 88 35 31 76 e-mail: fred@sobel.u-strasbg.fr ==================HamRadio: F1HFD (back to one letter prefix!)============= ------------------------------------------------------------------------------- From: Paul Leyland > Are you using PVM today? No, but I would like to be able to. > What are your application(s)? Computational Number Theory: integer factorization, primality proving, &c. > How many and what kind of machines have you used with PVM? A few DEC 5000 series, running MIPS Ultrix and DECathena > Any comments and suggestions for improvement are welcome. I struggled with pvm in an unsuccessful attempt to get it working under DECathena. It is *extremely* difficult even to get it to start working with Kerberos. When the tickets expire after a few hours, so do the pvm processors. File systems are not necessarily attached either 8-( Paul ------------------------------------------------------------------------------- From: Mike Fearn > Are you using PVM today? YES > What are your application(s)? MOLECULAR DYNAMICS SIMULATIONS > How many and what kind of machines have you used with PVM? FOUR HP9000/735 WORKSTATIONS ALTHOUGH WE WOULD LIKE TO UTILISE A LARGE i860 MULTI-PROCESSOR MACHINE (SEE COMMENTS BELOW). > Any comments and suggestions for improvement are welcome. A PVM ROUTINE TO RETURN THE VALUE OF MSGTAG OF RECEIVED MESSAGES WOULD BE USEFUL (WHEN USING WILDCARD MSGTAGS IN RECEIVE ROUTINES), TO SAVE HAVING TO SET UP PERSONALISED IDENTIFICATION LABELS TO SEND WITH MESSAGES. A PORT TO THE Meiko i860 COMPUTING SURFACE BOX WOULD ALSO BE VERY USEFUL. I'M SURE IT'S NOT TOO DIFFERENT FROM THE iPCS/860 BUT WE HAVEN'T GOT THE EXPERTISE/TIME TO DO THIS PORT OURSELVES. ------------------------------------------------------------------- | Mike Fearn Tel : UK (0684) 895371 | | D.R.A, Malvern, Fax : UK (0684) 896150 | | Worcs. WR14 3PS Email : hp003@signal.dra.hmg.gb | | OR fearn@hermes.mod.uk | | OR fearn%hermes.mod.uk@relay.mod.uk| ------------------------------------------------------------------- ------------------------------------------------------------------------------- From: laubie@etca.fr Hello, PVM Project Team ! Yes i'm using PVM today, version 3.0, between sparc stations, CM2 and CM5 ... (so about 4 machines ). My application is about subspace iterative methods in linear algebra. My remarks: - I've always messages like: "[t800c0000] netinput() pkt from t400b8 for t40040078 scrapped (not us)", that I can't eliminate, altough I think It's not very important... - I hope PVM will run soon on the nodes of the Connection Machine... - PVM does not know the name of the executable when the process is spawned from a shell prompt. It's not easy to recognize tasks with just the tids... - packing and sending should be realized by a parallel task, so the first task could compute while the second one should be responsible for the send. Waiting more... Best wishes, V. LAUBIE ------------------------------------------------------------------------------- From: markus@ii.uib.no I am using PVM today. I am using it for the development of parallel algorithms. I am using PVM3 with SUN4 workstations. In an earlier stage I used it (with mixed success) on the RS 6000. M. Hegland ------------------------------------------------------------------------------- From: Jalel Chergui Hello everybody, pvm-survey > Are you using PVM today? Yes, version 3.1.5 pvm-survey > What are your application(s)? Actually, PVM is installed in our Computing National Center machines. Today, it is used by a couple of users as an experimental sytem in view to get it accessible to all our scientific community (physic of plasma, chemistry, fluid mechanic, climatology, wheather broadcast, etc...) from the end of this year. In fact, our center will very soon include a cluster of RS6000 and a C98 from October this year and a massively parallel machine in 1994. pvm-survey > How many and what kind of machines have you used with PVM? Three kind of machines (as a virtual machine) IBM RS6000/520 DEC 3100 IBM 3090/600J under AIX/ESA pvm-survey > Any comments and suggestions for improvement are welcome. pvm-survey > Please send the information to pvm-survey@cs.utk.edu. pvm-survey > Your help is much appreciated. I would like to thank all of you for the very good job you are making concerning PVM project. Personally, untill today, i have solved all my PVM problems with patches and/or comp.pvm news group. Today, PVM is well adopted by many people. I think that in a near future users will need interfaces to develop applications with PVM. Sincerely yours, Jalel Chergui ********************************************** Jalel Chergui E-Mail: chergui@circe.fr CNRS/CIRCE Tel. : (1) 69.82.41.24 Bat.506 - B.P. 167 91403 Orsay Cedex FRANCE ********************************************** ------------------------------------------------------------------------------- From: Frank Bellosa > > Are you using PVM today? Yes, PVM 3.1.4 > What are your application(s)? CFD, theoretical chemistry, FEM in the field of mechanics > How many and what kind of machines have you used with PVM? CONVEX META-Series: 12 Processors HP735 Cluster: 7 Processors SUN4 Cluster: 20 Processors Motorola M88100: 80 Processors > > Any comments and suggestions for improvement are welcome. We are interested in the PVMINPLACE option for the pvminitsend function to decrease latency. What about a shared memory version for the KSR1 architekture (like TCGMSG)? Bye Frank -- --------------------------------------------------------------------- | Frank Bellosa | University of Erlangen-Nuernberg | | Martensstrasse 1| Department of Computer Science IV | | D-91058 Erlangen| mail : bellosa@informatik.uni-erlangen.de | | Germany | voice: (0049)-(0)-9131-857275 | | | fax : (0049)-(0)-9131-39388 | --------------------------------------------------------------------- ------------------------------------------------------------------------------- From: NE01@IBM3090.RZ.UNI-KARLSRUHE.DE Dear Colleague, I received your mail from Mon, 16 Aug 93. Here are the answers to your questions: - Yes, I'm using PVM today. - I'm working in parallelizing the program packages FIDISOL and CADSOL. FIDISOL and CADSOL are fully vectorized black-box solvers for arbitrary nonlinear systems of elliptic and parabolic partial differential equations on rectangular (FIDISOL) and arbitrary (CADSOL) 2-D and 3-D domains. FIDISOL and CADSOL have been developed at the university of Karlsruhe by the "Numeric research group". I'm a member of this group. - I've used 4 machines with PVM: 1 Sun 4, SPARCstation as master process and 3 Hp-9000 PA-RISC as slaves. Best wishes Matthias Schmauder Numerikforschungsgruppe Rechenzentrum der Universitaet Karlsruhe Postfach 6980 76128 Karlsruhe Germany Tel.: (+49) 721 / 608-4869 Fax: (+49) 721 / 32550 ------------------------------------------------------------------------------- From: Rudnei Dias da Cunha > >Are you using PVM today? Yes, and HeNCE as well. > >What are your application(s)? Iterative solution of linear systems. Programs are written with the SPMD model using Fortran77. > >How many and what kind of machines have you used with PVM? Up to 8 Sun SPARC 2 and 10 HP9000/400. > >Any comments and suggestions for improvement are welcome. I would like to suggest that a correction be made in the allocation of machines to processes -- currently the second process on the application is allocated to the machine where the application has been started. This is different from PVM 2.4 and causes some trouble when doing the validation and tests. Also, although this does not pertain to PVM itself: do you plan to include a library of routines to perform global operations? Those I found on BLACS are tailored to matrices as far as I can tell; in my application they are not useful. If you are planning to include such a library, can you give a rough idea when it will be available? My congratulations on such a good work! -- Rudnei Dias da Cunha Computing Laboratory University of Kent at Canterbury Canterbury, Kent CT2 7NF, United Kingdom ------------------------------------------------------------------------------- From: e8710@etlcom3.etl.go.jp (Asai Yoshihiro) -------------------------------- Are you using PVM today? -------------------------------- Yes. ------------------------------- What are your application(s)? ------------------------------ Large sparce matrix diagonalization. Application to models in solid state physics and statistical mechanics. ------------------------------------------------------------ How many and what kind of machines have you used with PVM? ----------------------------------------------------------- Sparc2 and 10 up to 6 WSs. IBMRS6K up to 4 WSs. --------------------------------------- Any comments and suggestions for improvement are welcome. Please send the information to pvm-survey@cs.utk.edu. Your help is much appreciated. --------------------------------------- Depending on network status, I ever met communication hangup. It seems to me that when the network is very busy, applications on pvm can hangup, even when there is no "bug". Sincerely, Yoshihiro Asai e8710@etlrips.etl.go.jp ------------------------------------------------------------------------------- From: asa@viking.Lanl.GOV (Yoshihiro Asai) -------------------------------- Are you using PVM today? -------------------------------- Yes. ------------------------------- What are your application(s)? ------------------------------ Large sparce matrix diagonalization. Application to models in solid state physics and statistical mechanics. ------------------------------------------------------------ How many and what kind of machines have you used with PVM? ----------------------------------------------------------- Sparc2 and 10 up to 6 WSs. IBMRS6K up to 4 WSs. --------------------------------------- Any comments and suggestions for improvement are welcome. Please send the information to pvm-survey@cs.utk.edu. Your help is much appreciated. --------------------------------------- Depending on network status, I ever met communication hangup. It seems to me that when the network is very busy, applications on pvm can hangup, even when there is no "bug". Sincerely, Yoshihiro Asai e8710@etlrips.etl.go.jp asa@viking.lanl.gov ------------------------------------------------------------------------------- From: "Dr." Matthew "D." Cooper > Are you using PVM today? Yes. We have been actively developing/porting applications to our workstations for around 6 months now. > What are your application(s)? An assortment of quantum Chemistry applications including molecular dynamics and ab initio codes. > How many and what kind of machines have you used with PVM? Two small(ish) clusters of Hewlett-Packard workstations, a cluster of Silicon graphics Iris machines and a few SG power series multiprocessor worksations. > Any comments and suggestions for improvement are welcome. The most important feature that I would like to see added to the version of PVM which I am currently using (3.1pl5) is the DataInPlace method of `packing'. In the initialisation of the parallel processes used in our applications the amounts of data which must be transferred tend to be quite large (of the order of a few Megabytes) and so the packing requires a significant length of time. In most cases we will only be running our parallel jobs within the homogeneous environment of one type of cluster or the other and so data translation is not essential. In this case the time saved by not packing the data could make a significant difference in the overall performance of the parallel program. Matt -- =================================================================== Matt Cooper Theoretical Chemistry Group Department of Chemistry Telephone: (+44) 61-275-4681 University of Manchester FAX: (+44) 61-275-4598 Brunswick Street JANET: mbdtsmc@uk.ac.man.ch.hpf Manchester, M13 9PL Internet: mbdtsmc@hpf.ch.man.ac.uk United Kingdom =================================================================== ------------------------------------------------------------------------------- From: Wu Qian Yes. We are using PVM for a large parallel CFD software. PVM version software has better portability but demand fast network to complete its advantages. As we only has Ethernet linking about 10 fast SGI workstations. We have the following result: Small test case will always win on 1 processor because of the communication overhead. Large test case will have speedup on 2 processors but speedup will degrade at certain number of processors(at our case 4-6 maybe the best number to have any speedup). We are looking forward to have FDDI installed later this year and by then we can predict some impressive efficiency. Wu Qian Dept. Computing Imperiall College London SW7 2BZ ------------------------------------------------------------------------------- From: J.P.Hagon One of our research students (Mark Cusack) has been doing most of the work with PVM. Here are his comments: Jerry.Hagon@newcastle.ac.uk ------------------------------------------------------------------------ Here are my answers to the pvm questions. 1) Are you using PVM today? - I tend to use PVM on a daily basis-if that's what you mean. 2) What are your application(s)? - I use pvm for several applications within solid state physics:- i) Calculation of semiconductor superlattice energy band structure. -typically calculating energy levels at 2000 or so points within the superlattice Brillouin zone. The set of energy levels at one point takes approximately 20 mins to calc. on a single Hewlett Packard 710. ii) Determining the contribution to the nonlinear susceptibility (chi2) from each one of the 2000 superlattice sample points - makes use of the band structure information obtained in i). 3) How many and what kind of machines have you used with PVM? - have run application i) on roughly 100 hp710's,a sun sparc 1, and a sun sparc 2 simultaneously. I have a question. It does not seem possible to use pvm3 FORTRAN applications on an IBM RS6000 running the AIX 3.1 OS. The problem seems to be associated with the m4 macro compiler supplied with the operating system. Is this true, and is there any way around it? (NOTE ADDED BY JH: If I recall, things were ok with PVM 2.4 on rs6k with AIX 3.1.5) thanks, Mark Cusack. --------------------------------------------------------------------------- -- =========================================================================== Jerry Hagon | E-Mail: Solid State Physics Group | INTERNET : Jerry.Hagon@newcastle.ac.uk Dept. of Physics | Compuserve: 100014,2702 The University | Newcastle upon Tyne | PHONE: +44 91 2227380 NE1 7RU | FAX: +44 91 2227361 United Kingdom | =========================================================================== ------------------------------------------------------------------------------- From: unibsa@vega.uni-c.dk (Bjarne Stig Andersen) Dear PVM Project Team. 1. Are you using PVM today: Yes 2. What are your application(s): Finite Elements for 3D Solid Mechanics. 3. How many and what kind of machines have you used with PVM: Today only one, but my max. configuration consists of 4 SUN's, 2 RS/6000's and a single Silicon Graphics. I also have access to a cluster of 4 HP's. With kind regards Bjarne Stig Andersen UNI-C The Technical University of Denmark, Building 305 DK-2800 Lyngby. E-mail: unibsa@vega.uni-c.dk ------------------------------------------------------------------------------- From: THOMAS@ibf.rwth-aachen.de > > Are you using PVM today? not today, but in this week I will use it > > What are your application(s)? parallel combinatorical optimization algorithms to optimize metal forming processes > > How many and what kind of machines have you used with PVM? four IBM RS6000 workstations > I am using FORTRAN and compared to the C - library the FORTRAN library is not very comfortable. Except that there is not much to criticise. rainer thomas E-mail: thomas@ibf.rwth-aachen.de ------------------------------------------------------------------------------- From: eblair@swift.mitre.org (Eric Blair) Dear Project Team; WHO: I work for the Center for Advanced Aviation System Development, a part of the MITRE Corporation. It is a Federally Funded Research and Development Center that serves the FAA. APPLICATION: We are investigating the use of PVM (among other software) as a software platform for implementing a large distributed discrete event simulation of the national airspace system (passenger and commercial air traffic). MACHINES: We have a small network of SUN4 machines under Unix and ethernet; 13 total, mostly SPARC 1+ and 2s with two SPARCstation 10s (model 30). We have a SPARC 10 multiprocessor (model 512 - I think) on order. We would be real interested in being able to pass messages through shared memory on this machine without going through the standard Unix sockets stuff. Any suggestions or help? COMMENT: I really like your product. It's robust, easy to install, well documented, easy to use. I just wish it were faster. Regards, Eric L. Blair ------------------------------------------------------------------------------- From: Thomas Radke Hello PVM Project Team ! You asked me to answer some questions about my personal use of PVM. 1. Are you using PVM today? Yes. 2. What are your application(s)? We develop a parallel simulation environment for neural nets (derived from the SESAME program of the GMD - the German National Research Center for Computer Science). Some parts of this simulator are running on a parallel machine. I take PVM as a easy-to-use mechanism for exchanging messages between the parallel machine and a local workstation. 3. All workstations are SUN4 machines, the parallel computer is a Parsytec GCel-128 (at the GMD) and a Parsytec Multicluster MC-32-2 here at the University of Chemnitz. I am looking forward to your next PVM release. Good bye ! Thomas Radke ------------------------------------------------------------------------------- From: leo@cp.tn.tudelft.nl (Leo Breebaart) > Are you using PVM today? Not yet, but planning to any day now. > What are your application(s)? We are planning to use PVM as a target language for a compiler that translates a very high-level source parallel language. The application written in that language will be mostly numerical. > How many and what kind of machines have you used with PVM? A network of Suns. > Any comments and suggestions for improvement are welcome. (1) Make PVM available through the number 1 Internet data communication facility: ftp. (2) Release separate versions when patches are made, so that we know unambigously whether someone is running 3.1.4, or 3.1.2 or whatever. -- Leo Breebaart (leo @ cp.tn.tudelft.nl) ------------------------------------------------------------------------------- From: stoltze@fysik.dth.dk (Per Stoltze) > Are you using PVM today? Yes and no. I'm studying the manual, sketching things on paper, checking specs of cluster computers ... but not actually compiling any code with pvm in it. > What are your application(s)? The application I'll put pvm into will be a materials simulation package. The user interface is based on the X Window system, so we already support graphics in a heterogeneous network. The next step is to use pvm to cut down the response time for large simulations. > How many and what kind of machines have you used with PVM? I will start using pvm on two Sun SPARC 10's "real soon now". Sincerely yours Per Stoltze ------------------------------------------------------------------------------- From: ANDREAS SCHWIERSKOTT Dear PVM-Team, I was very glad to see that the PVM Team is interested in the usage of PVM. Here are the answers: > Are you using PVM today ? Yes, I do. I'm a student of cs. at the Fachhochschule of Regensburg/Germany. I'm still a (very) novice with PVM. > What are your application(s)? PVM is the subject of my thesis for diploma. I want to implement a mesh(?)-problem. > How many and what kind of machines have you used with PVM? I'm currently using PVM on SparcStations with SunOs. My basic configurations are 4 SparcStations, but I can use up to ~14 Suns I'm currently using PVM on SparcStations with SunOs. My basic configurations are 4 SparcStations, but I can use up to ~14 Suns of the university network. If Solaris will become a bit more stable in the next time of the university network. If Solaris will become a bit more stable in the next time I'll also try to include the Solaris-Pool with 10 SparcClassics in my configuration. As I heard from Prof. Gentzsch (professor at the FH Rgbg. and boss of GENIAS Inc.) who is in contact with one member of the PVM team, there is now a "standard" called MPI (messages passing initiative(?)) which will probably be added to PVM. I'm very interested in this project and I would be very glad if you could give me more information about MPI and your plans to include it in PVM. In my opinion PVM is a wonderful tool, especially because of it's easy way to use (as far as I can see). I think PVM has good chances to be a useful helper in the university community where people usually have the problem to find a machine which is not too much loaded, where they don't bother to much the person who still wants to work interactively etc... There are two things I'm missing in PVM. 1. There should be more assistance of PVM in doing a rollback if one of the machines in the cluster fails. 2. Dynamic reconfiguration is important for good performance if one machine gets heavily loaded and slows down all other machines. A bit help from PVM would make things much easier. Yours sincerely Andy Schwierskott ------------------------------------------------------------------------------- From: massimo@vnet.IBM.COM Yes I am making a large use of PVM. The applications I run are int the area of linear algebra a computational chemistry. Since I work for IBM I am using only IBM RISC/6000 workstations. Usually I run with 4-8 machines. Massimo ------------------------------------------------------------------------------- From: james@sst.ph.ic.ac.uk (James Kew) > Are you using PVM today? I'm still playing with it having got it recently. > What are your application(s)? I have a Monte-Carlo simulation of MBE (Molecular Beam Epitaxy) - although this does not have any easy places to attempt parallelism, I'm thinking about using PVM as a batching system to automatically distribute runs with different parameter files over a network of workstations. > How many and what kind of machines have you used with PVM? 7 Suns of various kinds; I also have access to 10 SGI Indigos but haven't used PVM on them yet. > Any comments and suggestions for improvement are welcome. Implement pvm_notify. Add a different architecture for suns running Solaris - at the moment I have to run SunOS binaries in compatability mode which is less efficient than running Solaris native binaries. Thanks for providing PVM to the community - your efforts are much appreciated. ___________________________________________________________________________ James Kew "This dog and I are incompatible. For a start I think he's IC, London the wrong star sign..." Letter, TVQuick Problems Page ------------------------------------------------------------------------------- From: Norman.Barth@cerfacs.fr > [...] > Are you using PVM today? Yes > What are your application(s)? Climate modeling with coupled ocean/atmosphere models. In this application the heterogeneous/distributed/parallelism computing component is *not* what is of major interest. Instead "code reuse" is probably the most significant aspect. I use PVM to couple two big codes; one ocean model, one atmosphere model. The codes were too complicated to imagine combining into a single program with any hope of success. Realize each code is the product of years (sometimes a decade or so) of development by graduate students, postdocs etc. Thats to say the code is complicated, tortured, and you don't want to touch its "innards" to much if you can avoid it. Using PVM, its possible to couple models with very few alterations to their source code. I coupled the first models here in France with 4 source code changes/modifications in each model. This work - the PVM side - is briefly described in the proceedings of the 6th SIAM Conference on Parallel processing for Scientific Computing, March 1993. Sharon Smith is the co-author. Having said that, I *am* getting ready to use PVM's powers to run the ocean model on a Cray in Paris, and the atmosphere model on a Cray here in Toulouse (about 850km south of Paris). Other initiatives will be displaying simulation output in real time on my workstation etc. But thats the for the future. > How many and what kind of machines have you used with PVM? Right now I use PVM on one machine: a Cray-2. Soon it will be on a Cray-90, and perhaps later this Fall, on a distributed Cray-YMP/Cray-2 configuration. > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. It turns out that Cray Operating System is such that system operators are *not* able to put the PVM coupled models on hold. Hence they either have to kill and resubmit the job, or live with the job despite their desire to adjust the batch queues etc. (Realize *all* significant jobs on a Cray always run in batch mode.) I've written to Peter Rigsbee at Cray about this, but so far he hasn't responded, and I've been too busy to follow up. If this could be solved, it would be nice. Should I mention that I've shifted to PVM3.1? By the way, I have put a "PVM coupler" in the public domain. Its available via anonymous ftp from ahab.rutgers.edu in the directory /pub/DC . There is already a group in Belgium which is going to use this code, and some other people in computational fluid dynamics. Please don't hesitate to ask if you have any questions. Thanks very much for your ongoing work on PVM. Its a great product. Best wishes, - Norman (Barth) CERFACS Toulouse, France ------------------------------------------------------------------------------- From: paul@caplin.demon.co.uk (Paul Williams) We are using PVM today. Our applications are varied - we build i860 and transputer based parallel machines, and also configure and support workstation clusters used for parallel processing (mainly RS/6000s). Most of the applications are customers are using are their own parallel codes, mainly for scientific modelling etc. In house, we have our own 4-8 i860 machine, which we plan to run PVM on, and various RS/6000s, SUNs, VAXes. We have just configured and installed a cluster of 15 RS/6000s (each >90 SPECmarks) connected via FDDI and running PVM and IBM's LoadLeveller product. Some questions for you, should you have the time to answer them: 1. I have heard rumours of optimization work for PVM being done for RS/6000s running FDDI. Is this rumour or fact? Any details? 2. Does Hence work with PVM3.1 yet? 3. Are there any public-domain useful PVM applications? Thanks, Paul Williams, Head of software development, Caplin Cybernetics Ltd, London. ------------------------------------------------------------------------------- From: bhughes@ajax.mitre.org (Joseph B. Hughes) PVM Project Team - I'm not sure if you'll get feedback from others on this project, but here goes... (I'm not sure who, on the project, is on this mailing list) * Are you using PVM today? Yes. * What are your application(s)? We are using it to pass messages between "precincts" of a vehicular traffic model of traffic traveling through a town. The model is THOREAU, a testbed for simulating the results of various Intelligent Vehicle Highway Systems (IVHS), or communication and control technologies, for individual cars and traffic lights, etc. We divide the town into "precincts" and assign each precinct to a processor. The simulation is synchronized with barriers, and passes information about vehicles and availability of space on roads at precinct boundaries. * How many and what kind of machines have you used with PVM? We use a heterogenous system of SUN4 machines (mostly SPARC IPX's). Thank you for your efforts in making this package available to everyone. Sincerely, Brian Hughes Intelligent Vehicle Highway Systems C3 Modeling and Simulation Dept. (W156) MITRE Corp. bhughes@jasper.mitre.org ps. It would be great, if it isn't already out there, to have the latest version of PVM available with all patches applied... ------------------------------------------------------------------------------- From: paci@cecam.fr (Emanuele PACI) > > Are you using PVM today? Yes > > What are your application(s)? For the moment I am testing the possibility of using PVM on a classical Molecular Dynamics code. I am not using PVM on large scale. Just making tests and benchmarks. I am looking an efficient way of making load balancing. > > How many and what kind of machines have you used with PVM? Four different architecture and nine machine on one cluster and 6 omogeneous machines (DEC Alpha 3000/500) on another cluster. Both clusters are still on Ethernet. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. I have no suggestion for the moment. I find the package very easy to instally. The documentation is also good. emanuele paci ------------------------------------------------------------------------------- From: "Christian P. Roberts" > > Are you using PVM today? I am not a primary user, but can give support to those who want to use it in their research. I have seen that 2.4.1 is installed on our Sun Sparcs and, when I have the time, I may see to that being upgraded to 3.?.? Our primary user has been Bill Pearson, in Biochemistry. You have probably contacted him at wrp@virginia.edu. > > What are your application(s)? See above. Pearson's application was one in which a set of protein sequences (10 - 100) is compared to a larger set of protein sequences (2000 - 10,000) and similarity scores are calculated. > > How many and what kind of machines have you used with PVM? A network of 12 Sun4/40 IPC's. I tested it on several rs/6000's, but we would not run it that way publicly because of our accounting policies. > Chris Roberts Information Technology & Communication cpr4k@virginia.edu University of Virginia (804) 982-4693 Charlottesville, Virginia 22903 ------------------------------------------------------------------------------- From: hyland@sun1x.res.utc.com (Robert Hyland) > > Are you using PVM today? Yes, 3.0 we have used 2.4 in the past. > > What are your application(s)? Primarily CFD flow solvers (we have 5 applications in various stages of completion/development). > > How many and what kind of machines have you used with PVM? We have one application which could run on 15 workstations. Our primary workstations are IBM RS6000 but we do have some HP 710s. > > Any comments and suggestions for improvement are welcome. 1) I am waiting for the ability to transfer data "in place" and "raw", this would save quiet a bit of cpu time. 2) The applications I a working on have interactive portions, therefore I want to start them on my own workstation and have the pvmd start only the remote pieces. Currently the pvmd starts the remote processes in a round robin fashion which may causes my local machine to get another process loaded on it even if there are unloaded machines in my pvm pool. I have been pleased with the product so far, I have not encountered any problems which prevent me from doing my work. I suspect that we will use these applications in pools of less than 10 machines but we may be interested in trying them on a integrated parallel machine with more processors in the future. We are also interested in running these pvm applications on non dedicated machines. This means we are looking for a product which can manage the issuing, suspension and migration of pvm jobs. If you are working on something like this or know of such a product we would be interested. We are looking a DQS and have used LoadLeveler from IBM but neither has all the features we are looking for. ------------------------------------------------------------------------------- From: Nic "Fulton." I used PVM about 1 month ago and plan to resume using it in the next month. I am trying to write a program to calculate ro-vibrational wavefunctions and spectra for small molecules of between 2 and 4 atoms. This involves a large amount of diagonalisation which I perform using an altered EISPACK routine changed first by a Intel employer and then by myself to run under PVM 3. I have also written an asynchronous receive routine using the signal facility for both C and Fortran programs (Fortran using C home-made library). This is still incomplete as some signals get lost and others arrive to early. I have been using between 5 and 30 RS-6000 320 machines connected over a standard ethernet. I believe a good improvement would be a reliable asynchronous receive routine. Hope this is useful, Nic Fulton. ------------------------------------------------------------------------------- From: rhd@Arco.COM > >Are you using PVM today? Yes. > >What are your application(s)? Currently just testing small algorithms, but goal is parallel reservoir simulation. Biggest barrier is the lack of a robust, efficient, sparse, iterative linear solver. > >How many and what kind of machines have you used with PVM? PVM installed on a network of RISC 6000's. Have run on six machines simultaneously. ------------------------------------------------------------------------------- From: <@vm.gmd.de:ZWW@DGAGRS2A.BITNET> Dear Colleagues, Thank you for your mail of 16/08/93. 1. Yes, I am using PVM today. 2. My main application is a statistics code based on random numbers. 3. My configuration consists of a CONVEX C3860 and three IBM RS6000. Best regards, Winfried Zwermann ------------------------------------------------------------------------------- From: 1. I am personally not using PVM today, but I am working with others who are (they also received your survey and will respond under separate cover.) THe last time I used PVM was about a year ago. 2. Applications: Interior point optimization codes for two stage stochastic linear programs and nested decomposition codes for multi- stage stochastic linear problems. 3. I've used up to 6 DEC5000/120s and 8 IBM RS/6000-320's. Thanks, Derek Holmes Industrial and Operations Engineering, University of Michigan Ann Arbor, MI, 48109-2117 dholmes@engin.umich.edu ------------------------------------------------------------------------------- From: "Bill Wichser" > Are you using PVM today? Yes. > > What are your application(s)? Finite element code, namely dynaflow. > > How many and what kind of machines have you used with PVM? SGI, IBM RS6000. Preliminary groups of 3. Will expand to approx 10 in the near future. -Bill ------------------------------------------------------------------------------- From: Pinakin Kanubhai Patel 1.Are you using PVM today? NO. 2.What are your application(s)? Groundwater flow (multiphase systems) 3.How many and what kind of machines have you used with PVM? upto 8 machines... IBM RS-6000 Pinakin for monitoring processes, collecting info for load balancing, debugging etc. in the distribution. Port XAB to PVM 3.1. Best regards, Anders Alund ------------------------------------------------------------------------------- From: rgb@phy.duke.edu (Robert G. Brown) > > Are you using PVM today? Yes, with great pleasure. It has greatly facilitated my calculations, even though I am using it only a a fairly primitive mode. > > What are your application(s)? I am running a large scale heat bath Monte Carlo study of the 3D classical Heisenberg model using pvm as a parallel base. I am using it to increase run times and statistics, not to increase lattice sizes. As a consequence I run it in master/slave mode where I farm out N slave copies (each with their own random number seed) of the simulation started with the same data, and accumulate their results in the master program. I have written the program to be robust and reasonably restartable, if any machine(s) but the master goes down. I'm currently working on making it fully restartable. Because the problem, thus implemented, is extremely coarse grained (especially for the larger lattice sizes where it is most useful), I can safely spread the task out through three departments (physics, chemistry and math) here at Duke soaking up otherwise wasted CPU at very low priority. > > How many and what kind of machines have you used with PVM? I run on a net of as many as 30 Sun 4 workstations of various flavours. Most of them are SS2's, IPX's or ELC's, but there is a small cluster of six SS 1+'s and two each 630MP's and SS10/30's. The master is an SS2, although I hope to upgrade it significantly in the next year. So far I have chosen to keep the architecture relatively homogenous, partly to aid in the binary development and reliability. > > Any comments and suggestions for improvement are welcome. There are a number of things that I think could be done to improve the product. One is to improve the console interface, which is too primitive to be of much use and somewhat buggy as well. The ps facility, for example, provides almost no useful information such as the load average of the target machine and the percent CPU and memory dedicated to the task. A much larger superset of BSD's ps (say, a minimum of the (lrugax,ww) flags) is needed. An rwho-less implementation of the ruptime command for the pvm hosts is also greatly desired, so much so that I have written a script that performs somewhat that function). Unfortunately, a script has no good way of knowing which hosts are actively participating, and there is no guarantee that rwhod will be running on any of the hosts. A "better" way of obtaining the information should be programmed in as a console command. The system is not necessarily robust under a host going down. I have seen examples where pvm can actually crash sundry hosts if I/O is interrupted somewhere in such a way that pvm's buffers all fill. I obtained a 25 MB core dump that way (a personal record) and sent you same several months ago. Right now there is now way to sanely use the host "speed" variable, which seems to be set by architecture in a most generic manner. This is unfortunate, as speeds on Suns alone span a decade, with a SS10/41 ten times faster than an SS1. It seems unreasonable to assign them all a speed of "1.0". This should be a user-settable parameter, preferrably dynamically with the console or in the startup file if run from a file. After all, even two machines with the same real architecture can have differing "speeds" if one is a public machine and has several jobs running and another is private and dedicated to the task at hand. Finally, it has proven to be fairly difficult to program a restartable job. I currently trap a signal and force a reread of a configuration file that tells the pvm master to restart a slave on a machine (perhaps because it crashed in mid run, perhaps because I want to improve statistics with more independent runs). This solution is workable, but it would be nice to have some better built in facilities for doing so, in particular because if any handshaking involved in the startup is interrupted in mid-interrupt (as it were) one risks the crashing of the master program in my case, which essentially terminates the run prematurely. Many of these shortcomings can, of course, be overcome by programming pvm itself or by writing a suite of supporting scripts or binaries, but a number of them should really be built into the program itself. As a really truly final suggestion, it would be VERY nice to have a degree of built in X11 support, including an X-based console. X11 will ultimately be the other half of pvm's network support (pvm handling job parallelism across a net, and X11 handling distributed graphics across the same net). Work should be started immediately to build a really slick interface for the whole package. This will come in very useful when you are ready to do a full-scale debugger. rgb ------------------------------------------------------------------------------- From: yarkhan@menkar.EPM.ORNL.GOV (Asim Yarkhan) Are you using PVM today? Yes. What are your application(s)? A statistical sampling application. How many and what kind of machines have you used with PVM? Up to 32 machines, mixed SUN4s and RS6Ks. ------------------------------------------------------------------------------- From: Fred Kus Are you using PVM today? we recently finished installing and testing pvm What are your application(s)? we would like to use pvm to run 'gamess', a quantum chemistry program and a monte carlo code. there are also plans to develop fem codes that use pvm How many and what kind of machines have you used with PVM? 16 sparc stations Any comments and suggestions for improvement are welcome. the monte carlo code we are presently using was obtained from ornl we have not been able to find out if a pvm'd version is, or will soon, be available. it would be very useful to have email addresses of researchers who are porting codes to use pvm, or are developing new codes Fred +----------------------------------------------------------------------+ | Fred W. Kus INTERNET: fred@mcmail.CIS.McMaster.CA | | Computing & Information | | Services PHONE: (416) 525-9140 ext.4160 | | McMaster University FAX (416) 528-3773 | | Hamilton, Canada L8S 4M1 | +----------------------------------------------------------------------+ ------------------------------------------------------------------------------- From: ra@ux12.rice.ucl.ac.be >Are you using PVM today? Yes What are your application(s)? CFD Numerical Simulation of Viscoelastic Flows ( as Polymers ) How many and what kind of machines have you used with PVM? I am using a 8 processor cluster of HP735 processors. -rajesh ------------------------------------------------------------------------------- From: GianLuigi Alari 1) Yes I am using it 2) Mainly an interactive simulation steering and tracking syestem for distributed simulations 3) we test on two machines (SGI & RS6K) but we use pvm on a number of machines mainly RS6K, SGIs and SUNs. ------------------------------------------------------------------------------- From: bath@npac.syr.edu (Balaji Thiagarajan) > Are you using PVM today? Yes I am using it for a research project here at Syracuse University > What are your application(s)? This application is related to a project sponsored by Rome Air-Force Laboratories It is a compute intensive application for Battle Management Command and Control Simulation. > How many and what kind of machines have you used with PVM? I have used PVM 2.4.2 on cluster of Sun and IBM workstations. We have demonstrated the heterogeneous capability of PVM. Was not successful in using PVM 3.0 on an iPSC. Is PVM now available on CM5 and nCUBES. ------------------------------------------------------------------------------- From: ra@ux12.rice.ucl.ac.be >Are you using PVM today? Yes What are your application(s)? CFD Numerical Simulation of Viscoelastic Flows ( as Polymers ) How many and what kind of machines have you used with PVM? I am using a 8 processor cluster of HP735 processors. -rajesh ------------------------------------------------------------------------------- From: Peter Valasek Dear PVM-Team Here are the answers to your questions: We just installed pvm3.1 three days before. We are now testing the special features and example-programmes -> very nice so long. We thing to use pvm to block-parallelisize our Quantum-Monte-Carlo programs, so that we could make use of different machines for independent routines ( calculate measurements during the time- propagation etc. ..) We use HP-UNIX machines HP9000/700-755.They are connected to a cluster of 12 machines , so that we could test different configurations for accelarion or best load-balance. There is an non yet understood problem with the master-slave programs. If there are for example 4 computer running a virtual machine and we tell master1 to start with say 16 slaves, it may be that not all answers arrive at the master. Master1 waits, but slave1 has finished !! The status with ps tells that there are still processes, but not running! Best wishes Peter Valasek ------------------------------------------------------------------------------- From: anne@hermes.chpc.utexas.edu > Are you using PVM today? > Yes. > What are your application(s)? Computational molecular biology (several applications). > > How many and what kind of machines have you used with PVM? 8 Suns or 2 IBM RS/6000. Also, I'm using the 4 processors of a Sun 690MP which runs SunOS. I think that's the only way to use several processors of that machine as long as it doesn't run Solaris. (For my applications, it doesn't make much sense to use more than 8 machines.) > > Any comments and suggestions for improvement are welcome. - I don't like that pvm usually runs the first process started with pvm_spawn on the same machine where the manually started process runs. - I miss the ps -o option from pvm 2.4.2 where it was possible to see on which machines the last job ran. - I installed pvm in our binary directory /local/bin, and I didn't want the user to copy files to the home directory or make symbolic links there. I had to change several absolute path names in the pvm source. It would be nice if they could be eliminated. Related to this, I had a problem with pvmgs, because pvm looks for it in the ep directories, but I think it should be together with the pvm binaries and not with the user's applications. - I really like the debugger. I use it frequently and it helps a lot. Regards, Anne Juelich (anne@hermes.chpc.utexas.edu) Center for High Performance Computing The University of Texas System Austin, Texas ------------------------------------------------------------------------------- From: usimap@sneffels.usi.utah.edu (Michael Pernice) Hello, Following are responses to the survey you sent me. > Are you using PVM today? Yes, almost daily. > What are your application(s)? I am doing research in domain decomposed preconditioners for nonsymmetric pdes, with eventual application to problems involving Navier-Stokes equations. I use PVM to prototype software that implements these methods and run on clusters of workstations. When the implementation seems reasonably well under control, I convert to run on a multicomputer to study parallel performance. I am also working with a student on the parallelization of a molecular dynamics code, and have worked with a colleague on the parallelization of his image processing code, both using PVM and workstation clusters. You may be interested in several other PVM-based applications that are in use/under development here: - lattice gauge theory calculations - ab initio quantum chemistry calculations - electrocardiography - seismic wave propagation (forward and inverse problems) > How many and what kind of machines have you used with PVM? Various models of rs6ks and Decstation 3000s. > Any comments and suggestions for improvement are welcome. I may be behind the curve on some of this (I'm about 2 weeks behind in reading comp.parallel.pvm, and haven't used pvm3 yet), but: - direct support for global operations (max, min, sum, product,...) for different data types. This becomes especially important if you want to use PVM on the nodes of multicomputers with different interconnection networks. - ability to examine the contents of message buffers - smarter scheduling by pvmd, taking into account contributions to load originating from processes not initiated by pvmd - checkpoint/restart capabilities. This is real tough, and I know there's been some work at CMU on this. But something simple, that allows pvmd to catch a signal, then notify processes it spawned that someone wants to shut down or suspend a PVM application. Then it becomes the user's responsibility to monitor the daemon for this info, and then notify the daemon of a good place to shut down (like, when no messages are pending). - ability to run more than once without restarting abmon and xab. In addition, it would be nice to stop the real-time display in xab and 'back up' to look at something more carefully. - implement PvmDataInPlace. In general, the overhead for sending messages is too high. Network-specific optimizations (e.g. for FDDI) would also be nice. I plan to convert to PVM3 soon and I look forward to using some of the new features, like multiple message buffers and debug hooks. ----------------------------------------+-------------------------------------- Michael Pernice | Utah Supercomputing Institute | 85 Student Services Building Bell-net: (801) 581-7708 | University of Utah Internet: usimap@sneffels.usi.utah.edu | Salt Lake City, Utah 84112 ----------------------------------------+-------------------------------------- ------------------------------------------------------------------------------- From: trfaulk@ca.sandia.gov (faulkner thomas r) In answer to your questions on PVM usage: -- I do not personally use PVM but I do provide applications support for the Paragon system at Sandia National Labs, Livermore. I have installed the 3.0 Paragon-vetted version of PVM on the Paragon. I have had several requests for PVM as it is much more widely used than the Paragon's proprietary NX messages passing system. -- The application areas include Materials Science, Quantum Chemistry and Combustion Research. -- The only system here at SNL/CA that I know to have PVM installed are a couple of SUNs and the Paragon. I have it installed on my SUN for testing purposes along with the 1.3 release of the HeNCE system. -- Tom Faulkner Parallel Systems Engineer Supercomputer Systems Division Intel Corporation Sandia National Laboratories, California ------------------------------------------------------------------------------- From: chchang@cs.Buffalo.EDU (Chun-Shi Chang) > > Are you using PVM today? No. > What are your application(s)? X-ray crystallography. > How many and what kind of machines have you used with PVM? 10 SUN-4's and 24 SGI's. ------------------------------------------------------------------------------- From: "Lan Yang" >Are you using PVM today? Yes, occasionally. >What are your application(s)? I use PVM for teaching parallel algorithms and parallel processing and let students do some projects on top of PVM. >How many and what kind of machines have you used with PVM? Five SunStations. >Any comments and suggestions for improvement are welcome. PVM runs slowly on our system and we'd like to have more reference papers and sample programs. Is any conference proceedings on PVM? Any workshop? ------------------------------------------------------------------------------- From: rodrigo@inf.ufsc.br (Rodrigo Augusto Campos) > Are you using PVM today? yes. > > What are your application(s)? A Branch and Bound library (distribuded, of course). > > How many and what kind of machines have you used with PVM? 11. 10 HP and 1 SUN. Rodrigo. ------------------------------------------------------------------------------- From: Clifton J. Williams Are you using PVM today? Yes What are your application(s)? Parallel research using FORGE90 systems. How many and what kind of machines have you used with PVM? SUN workstation SParc station 1+ ------------------------------------------------------------------------------- From: LLIU@DALVMIC1.VNET.IBM.COM 1. Yes I am using pvm today. 2. My applications are in petroleum industry. 3. Typically I use 1-8 nodes RS6K. ------------------------------------------------------------------------------- From: Lars Ramfelt > >Are you using PVM today? Yes >What are your application(s)? Network simulation. >How many and what kind of machines have you used with PVM? Usually between 10 and 80. Most SUN4 but sometimes we also use DEC Alpha, SGI 4000, HPPA. >Any comments and suggestions for improvement are welcome. Thank you for a very good tool! - - /Lars ------------------------------------------------------------------------------- From: Thomas E Michener Dear "The PVM Project Team" I have PVM on my Cluster of HP 9000/735 computers. However I have only used it once to determine the the communication performance. We will be running a parallel version of our CFD code TEMPEST in the near future, but not with PVM. The lack of a performance increase in communication when as we upgrade to FDDI, has resulted in us choosing different software. Different staff members at Battelle (PNL) have tried PVM on SUN's, IBM's, and the HP's. When the communication is improved to take advantage of FDDI, then I will try using PVM again. Tom Michener Senior Research Engineer K7-15 te_michener@pnl.gov _______________________________________________________________________________ From: jpallis@rosarita.engr.ucdavis.edu (Jani Macari Pallis) Are you using PVM today? Not at the moment. We are waiting for fortran to be installed on some new systems. How many and what kind of machines? We have used them with IBM RISC, SUN4 and will be using them with some DEC systems (Numbers respectively 2,4, 20) Applications... Computational fluid mechanics, and in particular aerodynamics problems. We have done both space shuttle and Vertical takeoff work. And will be doing hypersonic vehicles shortly. Comments...I wish I could be in contact with other users. We are in contact with the NASA Ames folks who use PVM. I wish there was some job scheduling with PVM. Thanks very much, Jani Pallis ------------------------------------------------------------------------------- From: CYRANO@vms.cis.pitt.edu Dear PVM Project Team, I am just beginning to use PVM, and am interested in using it to implement a wavelet transform algorithm. So far my experience with pvm has been limited to NeXT machines. So far, so good! Thanks for a great tool. Regards, Alan ------------------------------------------------------------------------------- From: Matthew AYM Au >Are you using PVM today? I have not used PVM in about 2 months. My main job was to install it on our system. >What are your application(s)? The only application I've written is a prime number generator just to play around with parallelism. I did not have a chance to write an infinite precision package so it won't find very large primes. Beyond this I have not actively persued writing PVM code. >How many and what kind of machines have you used with PVM? I have used PVM on the following. Cray YMP2/216 Convex C-220 SUN 6/690 6 - SUN Sparc 1+ 3 - SGI workstations 4D/85 R3000 Indigo Crimson Some improvements could be made in the way in which PVM is installed on the sytem. It would be nice to type "make depend" and "make install" to install PVM. Manual (man) pages would be nice too. ------------------------------------------------------------------------------- From: "James T. Beaupre 1.617.258.1585" > >Are you using PVM today? Yes. We now use PVM in some 'production' applications and alos in some experimental work. >What are your application(s)? Mostly Fortran codes such as LSCM from Penn State used in design for naval architecture. We are experimenting with some CFD codes and creating a library of basic numerical routines which use PVM. >How many and what kind of machines have you used with PVM? Currently we use PVM on 4 DEC alpha's (expanding to 8), 16 DEC mip's (5000 series), and 2 SGI mips's (Crimson's.) >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. Comment: Keep up the good work, read the newsgroup, and best wishes to everyone with the PVM project team. I'll be seeing some of you again at SC'93 in Portland. ------------------------------------------------------------------------------- From: "Dr. A. Gaber Mohamed" > Are you using PVM today? Yes. > > What are your application(s)? > 1- Particle simulations as a part of the Numerical Tokamak Project (a DOE HPCC project). 2- Matrix algebra. > How many and what kind of machines have you used with PVM? > 4 RS/6000 580s. ------------------------------------------------------------------------------- From: fang@cs.wvu.edu (T. Fang) > > Are you using PVM today? I don't use PVM everyday. > > What are your application(s)? Following is a paper that we are trying to write. We hope implement most of the features on PVM. Software Portability and Reusability in Distributed Systems -- Distribution of ADT's Tao Fang Abstract In our method, software components(ADT modules) are distributed across different machines. In this way, we can have a function/procedure to be executed at the most appropriate machine(important factors affect the decision such as speed of cpu, peripherals). This way, every operation in an ADT is an RPC if the ADT resides at other machine. We need to extend RPC to allow passing by reference. We argue that parameter passing mechanism is an important issue in distributed systems as it increases performance, especially in distributed systems. We extend the idea of Nilpotent Template to distributed systems. Since ADT's are sequential programs, we use several mechanisms of parallelism extraction to get better performance. In distributed systems, for resons of either performance or partial failure tolerance, it is also important to have the ability to move an ADT from one machine to another machine or replicate ADT's. These involve translation of representations in heterogeneous systems. We propose an representation independent way to translate representations of ADT's. To get fine grain parallelism, we also need distributed data structures, especially dynamic structures. If we have distributed virtual memory system, the above mechanism will be easy to realize. However, we should still consider the performance of distributed virtual memory Of course, the above mechanism may be transparent to programmers. We need to provide clean, yet efficient interface. > > How many and what kind of machines have you used with PVM? Right now, we are using several machines, most of them are SUN4, DEC Station. If it is necessary, we will use CM-5 and more other work stations. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. Do you plan to extend PVM to other operating systems, such as VMS? -- Tao Fang Dept. Computer Science | Tel: (304)-293-4447 (Home) West Virginia University | e_mail: fang@cs.wvu.edu ------------------------------------------------------------------------------- From: beck@amb4.ccalmr.ogi.edu (Bradley Carl Beck) > Are you using PVM today? Yes. > What are your application(s)? Most of the applications I have written were for a parallel programming class. The applications included: matrix inversion, solution of ODE's, and solution of PDE's by multigrid, finite element, and relaxation techniques. My future plans are to expand my work in fluid flow/transport modelling using a parallel finite element solution technique. > How many and what kind of machines have you used with PVM? HP7?? RS6000 DECstation 5000 --- Bradley C. Beck Voice : (503) 531-7986 Oregon Graduate Institute Internet: beck@amb4.ccalmr.ogi.edu of Science & Technology (OGI) ------------------------------------------------------------------------------- From: davidr@george.lbl.gov (David Robertson) >Are you using PVM today? Yes. >What are your application(s)? Visualization applications. >How many and what kind of machines have you used with PVM? Most at once: 16 Types of machines: Sun's, Cray's, CM-2, DEC's I have tried using it with the MasPar front end and the CM-5 front end and have had problems (these are still unresolved but on the back burner). David Robertson Lawrence Berkeley Laboratory ------------------------------------------------------------------------------- From: winchell@bohr.physics.upenn.edu (David Winchell) Hi- We are currently using PVM to develop a distributed-processing data analysis package for high-fold nuclear spectroscopy data. We are running on five SGI Iris Indigos. Regards, David Winchell Physics Department University of Pennsylvania ------------------------------------------------------------------------------- From: sondeen@ISI.EDU > > Are you using PVM today? yes, occassionally, when I have to recompile tons of files > > What are your application(s)? a "parallel make" facility (main routine parses the output of "make -n" and farms out the cc compiles to all available machines) > > How many and what kind of machines have you used with PVM? sparcs and Hp 730's > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. more examples would be nice (of more advanced features) ------------------------------------------------------------------------------- From: tobita@marc.co.jp Dear PVM Project team: Thank you for including me in your list. Here are my answers. > Are you using PVM today? Yes! I'm a member of PVM study group, whose leader is Tokyo Electron Co.,LTD. which is a distributer of Convex supercomputer in Japan. This group is a branch study group of Scalable Computing Working Group, led by Convex Computer Co.,LTD. in Texas. > What are your application(s)? The applications I'm working on are MARC MARC/Linear MENTAT. MARC is a non-linear finite element analysis program. MARC/Linear is a linear finite element analysis program. MENTAT is a pre/post processor for MARC and MARC/Linear. We have added the functionality of PVM to MARC/Linear, but this is just for a study, yet. > How many and what kind of machines have you used with PVM? I'm in charge of distributing PVM to each machine that we have. Here are machines that I installed PVM on. (PVM 2.4.1 base) machine os fortran num of machines Sun SPARC 4.1.1 1.3 1 Sun 3 4.0.3 1.2 1 NEC EWS4800/220 5.1 5.1 1 HP9000/715.735 9.01 9.01 4 IBM RS6000/220 3.2.2 2.3 2 IBM RS6000/520 3.2.1 2.2 1 DG AV310 5.4.1 ? 1 (PVM 3.1 base) machine os fortran num of machines Sun SPARC 4.1.1 1.3 1 Sun 3 4.0.3 1.2 1 HP9000/715.735 9.01 9.01 4 IBM RS6000/220 3.2.2 2.3 1 > Any comments and suggestions for improvement What do you say to use shared memory for local communication, like mmap() or something like that, between user program and pvmd or user program and user program on a same machine? With this, I guess, we can send messages as fast as task to task communication. (Then we don't need task to task functionality.) I know that same kind of approach has been taken on some multiprocessor machine. But as to single processor machine, I don't know. I also put this mail forward to my aquaintance, who is in charge of MARC/Linear development. And we are really glad, if you keep in touch with the users like us, about new version release, user's group meeting and so on. With Best Regards, Akio Tobita Nippon MARC Co., LTD. 4F Daiichi Seimei BLD. 2-7-1 Nishi-Shinjuku, Shinjuku-ku, Tokyo, Japan. ZIP code 163-07. phone. +81-3-3345-0181 fax. +81-3-3345-1529 ------------------------------------------------------------------------------- From: loebel@sc.ZIB-Berlin.DE (Andreas Loebel) Dear PVM Project Team, we are using PVM3.1.5 this time. We develop several parallel branch&cut codes for combinatorial optimization problems. At a first stage, we want to parallize the separation of the cutting plane part and the generation of improved bounds. Some colleague of our department 'Symbolic' also wants to use and test PVM. This department develops effective, applicationoriented procedures for the processing of symbolic data structures. the emphasis is on algorithms in computer algebra, togehter with symbolic-numeric procedures and expert systems. the department is involved in the advancement of the computer algebra system Reduce. In our department we use PVM on a cluster of SUN4, SUN4SOL2, HPPA workstations. For our department symbolic, I have also installed PVM for the architectures ALPHA, DGAV, PMAX, RS6K. Some Comments: - The default of PVMDFILE's location shoud be $PVM_ROOT/pvm3.1/lib/pvmd. I had the same problem with the debugger call. - Because different people prefer different debugging tools, you should give them a possibility to use different debugging tools. I have solved this by setting the debugger2 call in debugger to `exec xterm -title "${hn}:$argv[1]" -e $HOME/pvm3/lib/debugger2 $argv`. So any user of PVM can use his favorite debugging tool. - In archcode.c you have forgotten to define the archtype SUN4SOL2. (but I think you already know this). Best wishes, Andreas Loebel ------------------------------------------------------------------------------- From: peter wark Are you using PVM today? YES What are your application(s)? Vehicle Routing How many and what kind of machines have you used with PVM? 8 to 16 SUNs Any comments and suggestions for improvement are welcome: 1. Many new users appear to have problems recovering when the system crashes. It would be good to have an automatic clean-up facility. 2. It would be good if one could specify that tasks are NOT to be spawned on the machine from which the spawning request originates, without explicitly saying where tasks are to be spawned. 3. I have found PVM to be a very useful tool in my work, and easy to learn and apply once I learned to recover from initial crashes. Peter Wark Department of Mathematics and Computing University of Southern Queensland Toowoomba Qld 4350 AUSTRALIA email: wark@usq.edu.au ------------------------------------------------------------------------------- From: rehmann@serd.cscs.ch (Rene M. Rehmann) > >Are you using PVM today? > Yes >What are your application(s)? > -- Parallelization of isosurfaces computation of molekules on SGI and SUN Workstations. -- Parallelization of color correction algorithm for printing true-color images on our Canon-Colorlaserprinter -- Teaching environment for a Summer Students Course >How many and what kind of machines have you used with PVM? > About 6 SGI Workstation and about 40 SUN Workstation. Best wishes Rene Rehmann ---- Rene Rehmann phone: +41 91 50 82 34 Section for Research and Development (SeRD) fax : +41 91 50 67 11 Swiss Scientific Computing Center CSCS email: rehmann@cscs.ch Via Cantonale, CH-6928 Manno, Switzerland ------------------------------------------------------------------------------- From: borresen@dme.epfl.ch (Bjarne BORRESEN) > > Are you using PVM today? Only initial expermiments to learn the system. > > What are your application(s)? The goal is to apply pvm with an multi-block solver for fluid mechanics (CFD). > > How many and what kind of machines have you used with PVM? 6 HP700 (715,735 and 755) -- O----------------------------------------------------------------------O | B\orresen, Bjarne jr. | | | | email: borresen@dme.epfl.ch | _O/ it is amazing how | | tel: +41 (0)21 693 3531 | / |/\ difficult a simple | | fax: +41 (0)21 693 3646 | / \_ problem can be | | smail: IMHEF/DME/EPFL,CH-1015 Lausanne | _/ | O----------------------------------------------------------------------O ------------------------------------------------------------------------------- From: Christian.Schaller@lrz-muenchen.de (Christian Schaller) > > Are you using PVM today? > I'm responsible for pvm at our Computing Centre(LRZ=Leibniz Computing Centre. We deliver the computing power for the Universities in the munich area) We are working with pvm for about a year now. The actual version is 3.1. It is installed on a number of different machines. > What are your application(s)? > I only did some testing of pvm, like communication between our different machines. I know that some people at the Technical University Munich are using my installation. Among these are: - Institute for Applied Mathematics: Parallel Implementation of Multiple Shooting for Boundary Value Problems in ODEs. (Dr. Kugelmann) - Prof. Zenger: Simulation of Semiconductors > How many and what kind of machines have you used with PVM? > Cluster of SUNs, Cluster of HPs, CRAY Y-MP8, CRAY Y-EL, KSR1 I hope the informations are usefull. Will the collected information be summarized in a paper? I'm interested! Regards Christian Schaller -------------------------------------------------------------------------------- Christian Schaller email: schaller@lrz-muenchen.de Leibniz-Rechenzentrum phone: +49-89-2105-8771 Barer Str. 21 fax: +49-89-2809460 80333 Munich Germany -------------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Gernot Reisinger > Are you using PVM today? No. PVM may be a basis for a project starting in october '93. > What are your application(s)? Parallel, distributed simultion of discrete and continous models (DEVS, DESS. DEV&DESS). ----------------------------------------------------------------------------- Dipl.-Ing. Gernot Reisinger EMail: (InterNet) gr@cast.uni-linz.ac.at Dept. of Systems Theory Johannes Kepler University of Linz - Austria/Europe ----------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Adriaan Joubert > > Are you using PVM today? Yes. > > What are your application(s)? Load flow analysis in power systems (solution of large nonlinear systems of equations derived from networks) Financial modelling > > How many and what kind of machines have you used with PVM? > DEC-Alpha DEC-5000/240 DEC-5000/200 with an 8K DECmpp SunClassic > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > XDR encoding is killing pvm. More options and versatility for user-defined encoding is needed. If e.g. only the lowest 1 or 4 or 8-bits of an integer are needed there must be a faster way of doing the encoding and decoding. Thanks for a great system. We appreciate it, as we have been writing some of the RPC stuff ourselves -- what a nightmare! Adriaan Joubert ------------------------------------------------------------------------------- From: kenneth@erin1.ucd.ie yes alpha (Digital)/S.G Indigos, 7- and more. Now trying to integrate MaSPar 16K mac hine. Chemical/Biophysics-Biochemistry/Complex materials..polymers etc. ============================================================================ y= = Professor Kenneth A Dawson Tel +353-1-7062447 = = Department of Chemistry Fax +353-1-2837873 = = University College Dublin = = Belfield, Dublin 16 E-mail Kenneth@erin1.ucd.ie = = Ireland = ============================================================================= ------------------------------------------------------------------------------- From: SCHNEIDER@EMBL-Heidelberg.DE >> Are you using PVM today? Yes, I use PVM version 2.4. I will move to PVM 3.x soon. >> What are your application(s)? fast and sensitive protein database searches. Related with genome projects, grand challenge.... >> How many and what kind of machines have you used with PVM? I use a mixed workstation cluster consisting of maximum: a SGI 480 (8 processors), 5 SUN's (SPARC-10), 5 DEC-station 5000 in addition I used PVM 2.4 on an Alliant FX2800 (16 processors) soon (I hope) I will switch from PARIX to PVM on a Parsytec GCel (1024 * T805 transputer) and a Xplorer-station (16 * T9000 transputer). Reinhard Schneider ----------------------------------------------------------------------------- ***** ##### # # ##### # Reinhard Schneider ******* # ## ## # # # European Molecular Biology Lab. ********* # # # # # # # # Protein Design Group ********* ##### # # # ##### # Meyerhofstr. 1 ** ****** # # # # # # 6900 Heidelberg, Germany ******* # # # # # # Fax: +49 6221 387 517 ***** ##### # # ##### ###### schneider@EMBL-Heidelberg.de ----------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: flebbe@pluto.tat.physik.uni-tuebingen.de (Olaf Flebbe) Hi, > Are you using PVM today? It is used experimentally until now, but will evolve. > What are your application(s)? Optimazation of Biomechanical Systems with Genetic Algorithms. [ There are plans for Astrophysical Simulations ] > How many and what kind of machines have you used with PVM? 9 Silicon Graphics Indigo R3000 and R4000 and a Powerstation. (Demos run additionaly on some decstations) Cheers Olaf -- Olaf Flebbe, Theoretische Astrophysik Tuebingen Internet: flebbe@tat.physik.uni-tuebingen.de ------------------------------------------------------------------------------- From: CCG0521@v2.qub.ac.uk To: The PVM Project Team From: Parallel Computer Centre, The Queen's University of Belfast. Currently we are running PVM over a network of about 8 or 10 suns. We have also included a Cray Y-MP EL and are planning to add in HPs. DEC 3100s and an SGI. We are running applied maths applications over the PVM network. Hope this helps! +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Richard McConnell, Tel . +44 232 245133 ext. 3012 | Parallel Computer Centre, Fax. +44 232 230592 | The Queen's University of Belfast, E-Mail RK.MCCONNELL@QUB.AC.UK | Belfast BT7 1NN, N.Ireland, U.K. +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ------------------------------------------------------------------------------- From: elliott@risky.dcrt.nih.gov > > Are you using PVM today? We are not using PVM at this time but are planning to offer it as a tool in the near future on our cluster of Suns and RS/6000s. > > What are your application(s)? Protein folding, sequence analysis, molecular dynamics. -- Elliott Alterman eba@cu.nih.gov National Institutes of Health eba@nihcu.bitnet Building 12/2241 Bethesda MD 20892 ------------------------------------------------------------------------------- From: messina@isoar.df.unibo.it (Antonio Messina) > Are you using PVM today? Yes, mainly testing different applications at the moment > What are your application(s)? ray tracing particle mesh codes for galaxy formation and large scale structure of the universe > How many and what kind of machines have you used with PVM? #1 SGI GTX310 #1 SGI Indigo Elan #1 SGI Personal 35 #1 HP 750 #12 Sun Spark #1 Cray YMP/464 In different combination to test network-configuration importance Best wishes Antonio Messina ************************************************************** A M M Antonio Messina AA MM MM Dip. di Fisica - Univ. Bologna A A M MM M via Irnerio 46 - 40126 Bologna - Italy A A M M M FAX: (+).51.351162, (+).51.247244 AAAAA M M Phone: (+).51.351162 A A M M e-mail: messina@isoar.df.unibo.it ************************************************************** ------------------------------------------------------------------------------- From: Camiel Severijns > Are you using PVM today? Not at this moment. > What are your application(s)? Have been thinking about using it for distrubited simulation of low-energy ion scattering. > How many and what kind of machines have you used with PVM? I have made some small patches to run it on Linux. These patches relate to the not yet complete implementation of sockets in Linux. > Any comments and suggestions for improvement are welcome. I would be interested in a C++ interface to PVM. With kind regards, -- | Camiel Severijns | Surface Physics Group, Dept. of Physics, Eindhoven University of Technology | Den Dolech 2, POBox 513 5600 MB Eindhoven, The Netherlands | Internet: camiel@surf.phys.tue.nl ------------------------------------------------------------------------------- From: tempus0!alex@nat.vu.nl (Alexa Doboli) } } Are you using PVM today? } Yes, I am using PVM today. } What are your application(s)? } I used PVM for some little applications which I shall use for the BEST SUMMER STUDENT COURSE. } How many and what kind of machines have you used with PVM? } I have used 6 SUN SPARC workstations. } Any comments and suggestions for improvement are welcome. } Please send the information to pvm-survey@cs.utk.edu. } Your help is much appreciated. } For the moment I don't have any comments , but if I shall have some suggestions I shall send them. Bye! Alec( Alexa Doboli) alex!tempus0@nat.vu.nl ------------------------------------------------------------------------------- From: Escaig Yves Yes, I am using PVM today My application is a structural mechanics program using finite element method, parallelized with a domain decomposition approach. I use PVM on a cluster of IBM RS6000 (5) to improve the performance of the program. I also use PVM between workstations and supercomputers (Cray, Convex) to create a distributed application (typicaly graphics on the workstations and calculations on the Supercomputers) Suggestions : 1) it should be possible to say in which directory the remote programs are to be executed (it is possible today if I always use the debugger option and do a "cd" be executing the program) 2) I wrote 2 scripts to launch automotically pvm : they read the pvmhost file, clean any previous pvmd daemon on all the machines and remove the pvmd.xx and pvml.xx from the /tmp directory before starting pvm. This couls be included in next release. Finally, thanks to Al Geist for answering my questions ... ------------------------------------------------------------------------------- From: sscott@mcs.kent.edu (Stephen Scott) > Are you using PVM today? > YES > What are your application(s)? > We are in the process of developing a new high-level programming paradigm on top of PVM called "Heterogeneous Associative Computing - HAsC". We are using PVM to provide the communications and control of the network of machines. > How many and what kind of machines have you used with PVM? > Initial work is being done on heterogeneous group of workstations: SUN4, SUN3, HPPA, RS6K ------------------------------------------------------------------------------ Stephen L. Scott Phone: 216-672-4004 x215 Heterogeneous Processing Research Group voice mail: x466 Kent State University Fax: 216-672-7824 Math Sciences Building email: sscott@mcs.kent.edu Kent, OH 44242 ------------------------------------------------------------------------------- From: scherr@fim2.fim.wpafb.af.mil (Stephen Scherr) > > Are you using PVM today? > yes, I am working on adapting an application > What are your application(s)? > Computational Fluid Dynamics > How many and what kind of machines have you used with PVM? > SGI, IBM RS6000, Sun I have 20 available, although I have only used 4 so far ------------------------------------------------------------------------------- From: moorej@chert.CS.ORST.EDU (Jason Moore) > > Are you using PVM today? YES > > What are your application(s)? So far, parallelizing a neural network program (at a large grain size) > > How many and what kind of machines have you used with PVM? About 15: Suns (SPARC 1s, mostly) and HPs (Apollo 700s) > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. I like it! I am a unix power-less user, but I was able to get it up and running with little difficulty. Other than that, it seems to do what it advertises! Jason -------- Jason Moore Internet: moorej@research.cs.orst.edu Department of Computer Science Bell-net: (503) 737-4052 Oregon State University "Anything worth doing is worth doing with a smile" ------------------------------------------------------------------------------- From: Laurent.Colombet@imag.fr (Laurent Colombet) Dear Colleague, Yes, I'm using PVM today. Applications using PVM: - Modelling propagation of pressure waves --------------------------------------- The aim is to simulate the propagation of pressure waves, artificially created (for example by an explosion), in an environment with heterogeneous density. A typical application is petroleum search. - Movement estimation in a sequence of pictures --------------------------------------------- The estimation consists in evaluating a movement vector for each pixel of the picture. To do that, the program uses a multiresolution pyramid (similar to those of multigrid methods). As the aim of this experiment is to parallelize the program at low cost, we therefore broadcast the pyramid without parallelizing its computation. - Star Modelling -------------- In astronomy, the formation of low-mass stars is one of the main field of research. From the observations, we know that these stars are associated with a large amount of dust located in a circumstellar envelope. It is believed that this envelope is the remnant of the star formation process, i.e., the collapse under the influence of gravity of a large molecular cloud. One of the effect of the circumstellar dust is to scatter the light emitted by the central star, thus producing the large observed reflection nebulae and the polarization of the light of these stars. Because they depend strongly on the dust distribution, models of the observable quantities (polarization and intensity) are useful to understand the star formation scenarios. Accurate models allow to evaluate the geometrical parameters of dust distribution. Unfortunately, no analytical solution exists for that problem and a numerical method must be used. A powerful one is the stochastic Monte Carlo method. With it, the events (scatterings) are independent to one another and the use of PVM allows us to achieve very good performances. - Medical Imaging and modelling on PVM network -------------------------------------------- Medical imaging * Image segmentation: we use a 3D active contour segmantation method faor having a mathematical representation of different anatomical structures. * Image registration: the purpose is to register multi-modularity images (CT, echography, IMR, SPECT, PET...) in order to help surgeons to perform operation accurately using all these information. * Compton effect modelling in SPECT: the possibility to register CT (tomodensitometry) i,ages with SPECT images (tomoscintigraphy) allows us to modelize the major corruption, i.e. the Compton effect and thus to improve the reconstruction. This intensive computation implies the use of distributed computing. Medical modelling * Simulation of a neuron lattice. Movement control pathology in humans such as the Parkinson's disease leads us to simulate the dynamical behavior of neuronal 2D lattice. Each lattice node, which may be seen as a neuron, is described by two time-dependent variables, namely, a potential (denoted V) and a recovery variable (denoted U). The dynamics of U and V is based on the well known integrate and fire neuronal behavior. The numerical simulation of the corresponding equations was carried out by using an implicit second order method. And our parallel team works on the parallelization of Algebra methods and FFT. For instance we have an implementation of bidimensional FFT algorithms with overlapped communications. --------------------------------------------------------------- Description of the network: (Remark: I give an access to Pr. Dongarra for the following network.) |-------------------------------------------------------------------------| | Observatoire de Grenoble | |-------------------------------------------------------------------------| | Computer | Model | Network | |-------------------------------------------------------------------------| | osug1.observ-gr.fr | RS6K/580| Ethernet or fddi with osug2, | |(osug1-fddi.observ-gr.fr) | | ccalc, ccalc1 and ccalc2 | | | | and Token ring with osugf1 | | | | osugf2 and osugf3 | |-------------------------------------------------------------------------| | osug2.observ-gr.fr | RS6K/560| Ethernet or fddi with osug1, | |(osug2-fddi.observ-gr.fr) | | ccalc,ccalc1 and ccalc2 | |-------------------------------------------------------------------------| | osugf1.observ-gr.fr | RS6K/320| Token ring with osug1, osugf2 and | | | | osugf3 | |-------------------------------------------------------------------------| | osugf2.observ-gr.fr | RS6K/320| Token ring with osug1, osugf1 and | | | | osugf3 | |-------------------------------------------------------------------------| | osugf3.observ-gr.fr | RS6K/320| Token ring with osug1, osugf1 and | | | | osugf2 | |-------------------------------------------------------------------------| | C.I.C.G. | |-------------------------------------------------------------------------| | ccalc.grenet.fr | RS6K/970| Ethernet or fddi with osug1, ccalc1| | (cfcalc.grenet.fr) | | and ccalc2 | |-------------------------------------------------------------------------| | ccalc1.grenet.fr | RS6K/560| Ethernet or fddi with osug1, ccalc1| | (cfcalc1.grenet.fr) | | and ccalc2 | |-------------------------------------------------------------------------| | ccalc1.grenet.fr | RS6K/560| Ethernet or fddi with osug1, ccalc1| | (cfcalc1.grenet.fr) | | and ccalc2 | |-------------------------------------------------------------------------| We have a network of 8 SUN4 (sparc 2 and 10) and 3 DEC Alpha workstations too. Suggestion: to give with PVM a Posix thread package to improve the communications overlapping for instance. If you want more explanations you can send me a mail. Best regards, Laurent ---------------------------------------------------------------------- Laurent COLOMBET LMC-IMAG Tel. : (+ 33) 76 57 48 64 Parallel team e-mail : colombet@imag.fr 46, avenue Felix Viallet or colombet@cicg-calcul.grenet.fr 38031 GRENOBLE / FRANCE or colombet@osug1.observ-gr.fr Fax : (+ 33) 76 57 47 54 ---------------------------------------------------------------------- ------------------------------------------------------------------------------- From: harsh@CAST.MsState.Edu (Harsh Anand) > > Are you using PVM today? > No. > What are your application(s)? > Ocean model evaluation techniques, scientific visualization I used PVM for course work only. > How many and what kind of machines have you used with PVM? > I used upto 5 machines, SUN Sparcs and SGI 4D/340GTXB > harsh@cast.msstate.edu Harsh Anand CAST, MSU Stennis Space Center, MS 39529-6000 (601)688-2562 ------------------------------------------------------------------------------- From: intevep!tony@conicit.ve (Antonio De Lilla Ext. 57735) > > Are you using PVM today? Yes, that is my working tool. > > What are your application(s)? Parallel and Distributed applications > > How many and what kind of machines have you used with PVM? I have been using only 20 machines, all of them: "Sun 4", and I am planning to use 300 machines. > > Any comments and suggestions for improvement are welcome. I have problems starting and halting pvm, sometimes I have to remove the file /tmp/pvmd.nn on every machine that I want to use. ------------------------------------------------------------------------------- From: kmbrantf@dal.mobil.com (K. M. Brantferger [Kenneth]) I'm currently using PVM at Mobil for reservoir simulation on a distributed network of RS6K's (1-8). I'm doing this in a SPMD fashion utilizing a data parallel code written for the CM2 and ported to the RS6K's using a F90 preprocessor. As for comments, everything with 3.0 is fairly straight forward with the exception of the Dynamic Process Groups, which admittedly I have not studied very thoroughly. Ken Brantferger Mobil E&P Technical Center ------------------------------------------------------------------------------- From: archie@gemini.unm.edu (Archie Gibson) Dear PVM Project Team, I have installed PVM 3.1 on some of the machines at the University of New Mexico Department of Mathematics and Statistics. I have used 1 RS6K, 1 SGI, and 6 SUN4 machines to run your master/slave example. In order to use PVM in my research in N-Body Scattering Theory, I need to write a "Pool of Tasks paradigm" in Fortran. Unfortunately, the only example which you give is the program xep written in C, and I do not know much C. I would really like to see an example program written in Fortran, where a master holds a queue and farms out tasks to slave programs as they fall idle. If you have such an example program, I would be very appreciative if you would send me a copy of it. Archie Gibson Professor of Mathematics Department of Mathematics University of New Mexico Albuquerque, New Mexico 87131 FAX: 505-277-5505 EMAIL: archie@gemini.unm.edu ------------------------------------------------------------------------------- From: bigbro@neptune.cmc.uab.edu (Big Brother) Are you using PVM today? Sorry, not yet. We definately plan to use it. What are your application(s)? Molecular mechanics How many and what kind of machines have you used with PVM? It will be SGI Indigos, IBM/6000s ------------------------------------------------------------------------------- From: stange@polly.phy.bnl.gov (Alan Stange) Hi, I am not at this time using PVM. I am thinking of switching because it does have a better environment than TCGMSG, my curent package of choice. My applications are basic monte-carlo integrations which parallelize trivially. I get a near linear increase in speed. I've only used PVM for test purposes on a SUN 10. Your welcome -- Alan Stange (stange@bnl.gov) ------------------------------------------------------------------------------- From: Claudio Fernando Resin Geyer Hello, I am a teacher at Universidade Federal do RGS, Brazil. My research theme (and also for teaching) is distributed and parallel programming. My students have installed PVM 3.1 (Sun Sparc net). I plan to use PVM in academic projects and courses. I would like to receive messages from PVM group like that you sent to PVM users with 3 questions. Thanks in advance. Regards, Claudio Geyer. ======================= Claudio Geyer. Universidade Federal do RGS - Instituto de Informatica A. Bento Goncalves, 9500 - Agronomia 91501 - Porto Alegre/RS Brasil Phone: +55 (051) 3368399 R6802 Fax: +55 (051) 3365576 E-mail: geyer@inf.ufrgs.br ------------------------------------------------------------------------------- From: carson@ocfkms7.llnl.gov (Larry Carson) Answers to your questions: 1) Yes, I'm using PVM today. 2) I am trying to solve the linear Boltzmann equation for neutron tranport. 3) I have used PVM on the IBM RS6000's and shortly expect to use it in a network of SGI Iris Indigo and SUN (3/60's and Sparc's) workstation. ------------------------------------------------------------------------------- From: rmarson@rp.csiro.au (Ralph Marson) > > Are you using PVM today? > While I have PVM installed I have not used it for my application prefering to look for ways to simplfy my algorithm. If this cannot be done then I will use PVM to provide the raw computational power. > What are your application(s)? > I am interested in tracking atmospheric fluctuations in data collected by an optical interferometer in order to improve the signal to noise of our astronomical observations > How many and what kind of machines have you used with PVM? > I have used PVM (while testing it out) with 1 Convex C220 + 4 IBM RS6000 Model 320 + 4 Decstations 3100 + 12 Sun Sparc 1's, 1+'s or IPC's > Any comments and suggestions for improvement are welcome. > Any way to reduce the learning curve both for general users and system installers would be appreciated. Interestingly our system manager walked in here ten minutes ago and took the PVM manual away to investigate the possibility of in a year or two replaceing the Convex C220 with a farm of Dec alpha's and using some software to automatically distribute the jobs the these machines. Thanks for providing the software. Ralph Marson Experimental Scientist CSIRO Division of Radiophysics ------------------------------------------------------------------------------- From: salkm@solomon.technet.sg (Lee Kuok Ming) > > Are you using PVM today? > yes > What are your application(s)? > Finite Volume, multi-block CFD solver > How many and what kind of machines have you used with PVM? > 4 IBM RS6000-350. ------------------------------------------------------------------------------- From: Steve Johnson > Are you using PVM today? Me, not currently. Probably this fall. Some of my colleagues are beginning to use it. > What are your application(s)? Groundwater transport. Data visualization. Molecular dynamics. > How many and what kind of machines have you used with PVM? About 8 Suns, 5 SGIs, and a Paragon. The SGI number will be growing to about 15, and we will be adding around 5 RS/6k's and an iPSC/860. > Any comments and suggestions for improvement are welcome. Only 1: Modify the compilation/installation so that pvm can be installed by a sysadmin in a public (e.g. /usr/local/bin) area. I spent a great deal of time hacking Makefiles to get this installed, and it wasn't a pleasant experience. - Steve Johnson, Texas A&M University ------------------------------------------------------------------------------- From: ASMNCHONG@v9000.ntu.ac.sg > Are you using PVM today? Yes > What are your application(s)? Image compression > How many and what kind of machines have you used with PVM? Five DEC 5000 (MIPS R3000) ULTRIX workstation. Regards Chong Man Nang cmn@ntu.ac.sg OR asmnchong@ntuvax.ntu.ac.sg ------------------------------------------------------------------------------- From: CLEMATIS@IMAGE.GE.CNR.IT Are you using PVM today? Not yet. I am planning touse it in the near future. What are your application(s)? Distributed simulation; graphics. How many and what kind of machines have you used with PVM? I plan to use PVM on a local area network of about ten workstations (Sun and HP). I will try to integrate in the network a transputer machine (with about ten transputers). Best wishes for the PVM effort Andrea Clematis ------------------------------------------------------------------------------- From: mirnaz@appli.mimuw.edu.pl (Mirek Nazaruk) > > Are you using PVM today? No, I'm not, but people in our Institute are. We have started with the PVM verison 2.4.1 and now we are running pvm version 3.0. Though we have had PVM for a year it wasn't used a lot. But now it seems to be. > What are your application(s)? The main fields of our interest are numerical algorithms. Currently people in our Institute start to implement additive Schwartz method for solving PDE. > > How many and what kind of machines have you used with PVM? We are running PVM on SUN's(4) and Evans & Sutherland graphic workstations(4). Best wishes, Mirek ---------------------------------------------------------------------- Mirek Nazaruk Institute of Applied Mathematics and Mechanics, Warsaw University Banacha 2, 02-097 Warsaw, Poland tel: +48+2+658-42-36 e-mail: mirnaz@appli.mimuw.edu.pl ---------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Adrian F Clark >> Are you using PVM today? Yes. We are in the early stages of developing a fairly large application based upon PVM. >> What are your application(s)? We are researching into the efficient processing of image data on both "traditional" parallel processors (e.g., transputer) and non-traditional ones. Our work to date has primarily been on transputer-based hardware, where we can make good use of data parallelism, and we are extending our research to look at networked systems. The interest in PVM stems from: o it provides a portable user interface to the underlying parallel system (it would be nice if it could be shown to run on systems other than Unix, such as VMS and Linux) o a number of workstation vendors are adopting it for their own multiprocessor systems What we are actually going to do is to develop an implementation of (at least some of) the recently-approved ISO standard for image processing. The intention is that it will be able to soak up any spare cpu time on a network of machines more or less transparently to the user, as well as being capable of use on multiprocessor and perhaps even truly parallel systems. >> How many and what kind of machines have you used with PVM? We are currently using HPPA, SUN4, SGI, and PMAX architectures, with several machines in each class. We are thinking of porting it onto a transputer system, a Parsys SN1000 running TransIdris. This would be an Idris, not raw transputer, port. >> Any comments and suggestions for improvement are welcome. 1. From my newly-initiated perspective, the biggest problem with the PVM is currently the documentation! The PVM 3.1 manual does not explain some things well, particularly for the new user or installer. There needs to be an overview that describes PVM and its associated tools (and it also needs to say which tools -- such as xpvm -- are now defunct). There are also (forgive me for saying so) a fair number of typographical errors in the PVM manual. And it needs an index. To make a constructive suggestion, if I could be provided with the source of the document, I would be happy to convert it to LaTeX (it looks like plain TeX at the moment), incorporate the figure that is currently an insert, eradicate as many of the typos as possible, add an index, and so on. I could also provide versions of the document that could be used with WWW (World-Wide Web) and info readers. I could even generate a set of Unix manual pages for those people without more modern documentation systems. 2. If I haven't got mixed up with a problem that disappeared with v3, loss of communication with the machine that initiated the VM is not handled. Addressing this problem would be a Good Thing. But I don't have a specific suggestion for improving it. 3. I believe the scheduling algorithm for processes to hosts is purely round-robin, not even taking into account the relative speeds of hosts: this is likely to cause an uneven distribution of load; it's certainly sub-optimal. Our approach is to hook the PVM library directly into a command-line interpreter, obviating the use of the PVM console, and to monitor, from the CLI, how much work each machine actually does: this lets us dynamically adjust the partitioning of work. When the code is working to my satisfaction, I'd be happy to release it. 4. The ability to read and write remote files via PVM calls would be useful, perhaps along the lines of Sunderam's talk at the PVM UG meeting. I know that one can bung any required data into work packets, but it is often much easier to modify existing serial code to perform remote file access than it is to pack and unpack packets of data. I hope these comments are of some help. Keep up the excellent work! Best wishes, ..Adrian Dr Adrian F. Clark JANET: alien@uk.ac.essex INTERNET: alien%uk.ac.essex@nsfnet-relay.ac.uk FAX: (+44) 206-872900 BITNET: alien%uk.ac.essex@ac.uk PHONE: (+44) 206-872432 (direct) Dept ESE, University of Essex, Wivenhoe Park, Colchester, Essex, C04 3SQ, UK. ------------------------------------------------------------------------------- From: efields@eng.clemson.edu (ernest fields) At present we are using PVM over 3 platforms, Sun, HP, and DEC stations. The types of applications are mainly vision with some work being done on simulating various type of architectures. Craig Fields ------------------------------------------------------------------------------- From: William Gropp I use PVM only as an available transport layer for Chameleon. Bill Gropp ------------------------------------------------------------------------------- From: ikelle@scrsu1.SINet.SLB.COM (Luc Ikelle) From: Luc T. Ikelle Schlumberger Cambridge Research High Cross, Madingley road Cambridge, CB3 0EL ENGLAND. Are you using PVM today? Yes, I am using PVM software from Gseit et al. (1993) What are your application(s)? I am runing it to solve the linearized inverse problem for 3D multioffset seismic reflection data. How many and what kind of machines have you used with PVM? I am using four IBM RS6000. ------------------------------------------------------------------------------- From: shamis@iris.umcs.maine.edu (James L. Fastook) glad to answer ur survey: 1) are u using PVM? yes, in a course on parallel programming, we use PVM on our local network, and APF (a parallel FORTRAN) on the Cornell IBM9000 (six processor). 2) What application? We do various standard CS-type algorithms, sorting, memmber of, numerical methods (matrix inverse, matrix transpose, integration), and others, all as examples of parallel programming, no production code. 3)How many and what kind of machines? 20 SUN workstations. thanks, and glad to be of assistance, your product is very good and most useful. keep me apprised of updates etc. jim fastook computer science university of maine orono, ME 04469 fastook@maine.maine.edu or above 207-581-3927 ------------------------------------------------------------------------------- From: mln@blearg.larc.nasa.gov (Michael L Nelson) >Are you using PVM today? yes. >What are your application(s)? development, troubleshooting, and benchmarking of NASA Langley's Workstation Cluster. >How many and what kind of machines have you used with PVM? most recently, sun IPX's, SGI IRIX's, and a IBM RS6000 Michael Nelson NASA Langley Research Center ------------------------------------------------------------------------------- From: als@rwdws.ctd.ornl.gov I am using pvm 3.0 on a network of Dec Alphas. I am trying to run a large groundwater transport model. ------------------------------------------------------------------------------- From: srao@ISI.EDU (Santosh Rao) 1) Are you using PVM today? Yes, but in a limited sense. 2) What are your application(s)? We are mainly studying the API implemented by the PVM library routines. We are not involved in writing applications using PVM. Our main goal is to integrate PVM with the Prospero Resource Manager (PRM). The Prospero Resource Manager provides a uniform and scalable mechanism for allocating processing resources in parallel and distributed environments. It uses multiple managers to reduce the complexity of resource allocation and efficiently utilize the system's resources, while presenting the user a single system image. For running parallel applications, we believe that the automatic processor allocation mechanisms of PRM complement the communication libraries of PVM very well. We are currently implementing a PVM library interface to the Prospero Resource Manager. This will enable users to run PVM application programs in the PRM environment with minimal modifications, and also take advantage of PRM's resource allocation facilities. 3) How many and what kind of machines have you used with PVM? Mainly 3-4 sparcstations. 4) Any comments and suggestions for improvement are welcome. We believe that the integration of PVM and PRM would be beneficial for both projects. We would appreciate your comments regarding this. Perhaps we could also collaborate in this effort. The PRM project is ongoing at the USC Information Sciences Institute, and a beta release is available. A paper discussing the high level design of PRM and some preliminary results is also available. It is titled "Resource Management for Distributed Parallel Systems", and was published in the proceedings of the 2nd International Conference on High Performance Distributed Computing (HPDC2). It is also available by anonymous ftp from the host prospero.isi.edu as /pub/prm/papers/prm-hpdc93.ps.Z. Thank You. Santosh Rao. srao@isi.edu. ------------------------------------------------------------------------------- From: evakylee@masc1.rice.edu (Eva Kwok-Yin Lee) Yes, I am using pvm on a network of SPARC stations. The application is to develop a parallel branch-and-bound code for mixed integer program. We are also planning to run it on networks of DEC machines. Eva Lee ------------------------------------------------------------------------------- From: stone@ece.nps.navy.mil (Leon Stone 12-93) I was asked by Professor S. Shukla to answer your survey request since I am currently using PVM in working on my master's thesis. PVM was used in a parallel processing course last quarter as a tool to help students get practical experience with parallel processing. PVM was used for two introductory labs and one final project. I am currently using PVM on a network of 20 SUN workstations. I am using it to demonstrate the benefits of parallel processing on various Navy specific applications. The send/receive overhead is too costly for some of the programs and pvmadvise does help, but it is "fragile" and any improvement in overhead cost would be appreciated. ------------------------------------------------------------------------------- From: harumi@rigi.tis.toshiba.co.jp Hi,I'm engineer on high performance computing and numerical analysis of Toshiba corporation. >>Are you using PVM today? Yes,I use always it recentry. >>What are your application(s)? I treat various applications. For example, I'm making simulation system for neutron transport using Monte Carlo Method. >>How many and what kind of machines have you used with PVM? Cray Y-MP, Sun4 >>Any comments and suggestions for improvement are welcome. >>Please send the information to pvm-survey@cs.utk.edu. >>Your help is much appreciated. I want to add facility of loar balance analysis and task scheduling like IBM "load leveler". Sincerely, TOSHIBA Corp. Total Information & System Div. 580-1,Horikawa-cho Saiwai-ku kawasaki-shi 210 JAPAN TEL.+44-548-2780 FAX.+44-548-8985 (MS.) Harumi Hasebe ------------------------------------------------------------------------------- From: tony@sparky.esd.mun.ca (Tony Kocurko) Dear Colleague: Unfortunately, due to a plague of alligators, I have not been able to do other than install PVM. Our proposed applications include 3-D reflection seismic processing and seis- mic tomography. Although I have only installed PVM on a pair of SPARCstations (a 1+ and a 2), it is slated for installation on a pair of SPARCstation 10/30s and on a bigger SPARCserver 1000. We also have an ageing Convex C-1 XL on which it will be in- stalled once a donation of new disks and extra memory are integrated. Regards, Tony Kocurko ------------------------------------------------------------------------------- Seismological Systems Manager e-mail: tony@sparky.esd.mun.ca Departmant of Earth Sciences akocurko@leif.ucs.mun.ca Alexander Murray Building - Room ER-4063 office: (709) 737-4576 Memorial University of Newfoundland fax : (709) 737-2589 St. John's, Newfoundland TELEX : 016-4101 MEMORIAL SNF Canada A1B 3X5 ------------------------------------------------------------------------------- From: fox@vulcan.nrlssc.navy.mil (Dan Fox) > > Are you using PVM today? Not on a regular basis, although I plan to begin again soon. > > What are your application(s)? We are the part of the Naval Research Lab developing the Navy's ocean nowcast/forecast systems. The eddy-resolving circulation models and high resolution near-surface mixed layer models will run operationally on a Cray YMP or C90 at one of the operational centers. Our development process is sometimes hampered by restrictions placed on these machines so quite a bit is actually done on Suns and Irises as well. Applications which are easily converted to run under PVM without doing too much "damage" to existing code (important since the code is ultimately aimed at a different machine which will not be running PVM) include: * optimum interpolation of irregular oceanographic measurements onto a regular grid. * acoustic raytracing * some of our basic image processing None of these codes has been fully tested under PVM yet however. At present, these are just planned applications. All these applications also involve no inter-process communication. They are "trivially parallel" in a sense. > > How many and what kind of machines have you used with PVM? So far, only a network of Suns (including Sparc 1, 1+ and 2; we now have several Sparc 10's which will be added next time I use PVM). The Sun network as of the start of the next fiscal year will be about 6 Sparc 10's, 10 Sparc 2's, and 10 Sparc 1's. Memory runs from 24 to 64 MB. This network is a significant fraction of the power of the Cray YMP our efforts are ultimately aimed at. (Note: we also have access to an RS6000 and a couple of Irises but these machines tend to be busy all the time so I have not tried to bring them into the pvm network yet.) > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > I wish there were more examples of various pvm'ized applications available (and not just the code, but perhaps also a document describing the approach. Here's the way XXX is done normally, and here's the PVM way of doing it. Perhaps a couple of variations: first, the minimal changes to the code so that at least some parallelization is achieved, then a complete rewrite so take maximum advantage of PVM. Perhaps a few flowcharts or algorithm descriptions.) Is there somewhere people might contribute codes perhaps? It seems like somewhere along the line people would have written PVM versions of multi-dimensional FFTs, convolutions, basic numerical algorithms (I've seen BLACS but this is just building blocks... perhaps someone could donate a linear systems solver or an eigenvalue package based around these tools). Overall I think your manual is very good. Having both C and FORTRAN examples together is great. We do all our scientific programming in FORTRAN but occasionally some public domain code in C comes our way that we'd like to be able to modify locally as well. Having examples of both in the manual is helping me get more comfortable with C... +----------------------------------------------------+ | Daniel N Fox fox@nrlssc.navy.mil | | Naval Research Laboratory, Code 7323 | | Stennis Space Center, MS 39529-5004 | | Telephone: (601)688-5588 FAX: (601)688-4759 | +----------------------------------------------------+ ------------------------------------------------------------------------------- From: jsun@ncsa.uiuc.edu (Jun Sun) > >Are you using PVM today? > Yes, I will. >What are your application(s)? > Simulation of Discrete particle movement. >How many and what kind of machines have you used with PVM? > I've using PVM on RS6000, HPPA, and now on CM5. >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. > In many fine grained applications, the communication in the major loop will probably kill the PVM approach becasue of the overhead. One useful technique I found before in PC was to overlap the computation and communication. I was told in some machines this is "hardwarely" impossible. Not sure if there are some ones possible, and not sure if the software will exploit that if it's possible, like some sort of "Direct Memory Tranmission" mechanism. ------------------------------------------------------------------------------- From: tremback@janus.atmos.colostate.edu I am using pvm today (even as we speak!). The application is an atmospheric simulation model (Regional Atmospheric Modeling Systsem- RAMS). We have used so far only RS6000's (up to 12 of them). Suggestions: It is rather difficult to get printed output back to the master machine when trying to find bugs, especially in Fortran. Even the techniques used in version 2 was more usable for debugging purposes. It would be real nice to standardize "xab". Has this been ported to version 3.1 and the RS6000 yet? - Craig Tremback ------------------------------------------------------------------------------- From: schu@rucs.faculty.cs.runet.edu Are you using PVM today? > Yes. What are your application(s)? >- characterization of parallel algorithms in general. > - as a tool for graduate students to experiment with parallel algorithms. How many and what kind of machines have you used with PVM? > SUN 4's (35) and SUN 10's (4) Any comments and suggestions for improvement are welcome. > Not at this point. > Thank you. ------------------------------------------------------------------------------- From: marquez@quantix.us.es (Antonio M. Marquez 455-7177) > Are you using PVM today? Yes. I'm currently using PVM on production jobs as well as I am continuing to parallelize programs using it. > What are your application(s)? HONDO8.4 from IBM. A quantum chemical generally available. I make contributions to it, both to the serial program and I did program the parallel version using PVM (that was last year). > How many and what kind of machines have you used with PVM? Some RS6K, DEC-3200 and others running Ultrix, and various CONVEX mini-supercomputers (C2 and C3) > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. For now I don't have any. I think is a nice and very general way of using a network of computers as a single multi-cpu parallel computer. You did a nice job. Best wishes, ------------------------------------------------------------------------------- From: "Kwok L. Chow" > > Are you using PVM today? yes > > What are your application(s)? Parallel numerical computation > > How many and what kind of machines have you used with PVM? more than 30 SUN workstations > > Any comments and suggestions for improvement are welcome. A nice X window interface for coding and debugging is preferred More accurate and easy-to-use timing mechanism is needed ------------------------------------------------------------------------------- From: naiman@mrj.com (Aaron Naiman) Hi! I am currently not using PVM, but I have in the past, and hope to in the future. My applications are mainly finite element analysis and numerical linear algebra. In general, I have been very pleased with the package and with the friendly, prompt and proffesional technical assistance I have received. Also, the time taken to package the software with detailed scripts (even if they are not always _just_ what I wanted, some of them being heuristic in nature) was refreshing and very much appreciated. My primary interest these days is in the CM5. Thanx, and keep up the great work. It seem like the PVM team is a good model for team work. [BTW(/suggestion), who's who in the PVM Team picture PostScript file?] Aaron Aaron Naiman | MRJ, Inc. | University of Maryland, Dept. of Mathematics | naiman@mrj.com | naiman@math.umd.edu ------------------------------------------------------------------------------- From: gaudet Hello ! > > Are you using PVM today? Not at the moment but in a near future ... > > What are your application(s)? Scientific computing in the mecanical fields. > > How many and what kind of machines have you used with PVM? Between 5 and 10 computers : Sun IBM HP. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. I think that a form of load balancing system could be very interesting. > Your help is much appreciated. Thank you and the very serious work of PVM. ----------------------------------------------------------------------- | Michel GAUDET | | Ecole Nationale Superieure des Mines de Paris. | | Centre des Materiaux. | | Enceinte SNECMA B.P. 87 91003 EVRY Cedex | | ATTENTION Nouveau Tel 60763048 | | gaudet@mat.ensmp.fr | -------------------------------------------------------------------------- ------------------------------------------------------------------------------- From: Jiri Navratil Dear Developers, I appreciate your interest about the (potential) users. Currently, I am trying to establish the team and create the Center for Intensive Computing which should be as a cluster server in our network CESNET (Czech Scientific Network) which provide services to the all Czech universities. We are in the beginning and therefore we are getting all information and experiences from the network. > > What are your application(s)? We expect several types of applications from the Chemical-enginnering Fluid-dynamic and from the Civil-engineering fields. There is also big interest to study these problems as a discipline in the Computer Science Department. > > How many and what kind of machines have you used with PVM? > The first several RISC machines (2xRS6000/375 and 1xHP9000/735) will be delivered in the end of this year. Best regards, Jiri Navratil, Computer Center of Czech Technical University, 166 35 Prague 6, Zikova 4, Czech Republic ------------------------------------------------------------------------------- From: Andrea OMODEO Well, first of all, plase forgive my bad english. I'm working for the Italian Electric Power Company and I've discovered PVM just a couple of months ago. We're working on the simulation of thermoelectrical plants and currently have a set of simulators running on a transputer network under Helios. In perspective, this application should be ported to PVM. As a second project, it will be interesting to port PVM under Helios (wich has a socket library). In our labs we have some RS6K, a SUN4 and we're buying an APLHA machine. From my little experence, i've nothing special to suggest you, but i wish you publish a little mor about PVM internal structure and workings. Thank you for your VERY GOOD work. Andrea OMODEO (agnelli@cdc835.cdc.polimi.it) ------------------------------------------------------------------------------- From: dza@pc-dza.info.fundp.ac.be (Denis Zampunieris) >> Are you using PVM today? Yes. (version 3.1) >> What are your application(s)? First, students have developped some academic projects using PVM, in order to exercise their parallel programming knowledge; it was mainly clients-server applications. Second, we are trying to parallelize two research programs: a Prolog programs static analyzer, and an on-line rules based audit-trail analyzer (Asax, in cooperation with Siemens Nixdorf). >> How many and what kind of machines have you used with PVM? The original configuration was formed by an Ethernet LAN of 10 Sun workstations. It is still considered sufficient for our works, and we have no plan to extend it. Regards, B. Le Charlier, D. Zampunieris Institute of Computer Science University of Namur 21, rue Grandgagnage, B-5000 Namur Belgium. ------------------------------------------------------------------------------- From: andy@scp.caltech.edu (Andy Fyfe) > Are you using PVM today? Not quite yet, but within the next couple of weeks I'd guess. > What are your application(s)? We want to simulate the J-machine (perhaps more accurately its parallel computing ideas), and already have simulators for both "real" machines (the Touchstone Gamma), and for "virtual" machines (we have one based on the Cosmic environment, and another for mulitprocesses on a single SysV machine). PVM will provide the basis for another simulator. > How many and what kind of machines have you used with PVM? Sun 4. If others are interested in our code, they may want to use the pvm-based simulator in other environments. --Andy Fyfe andy@scp.caltech.edu Scalable Concurrent Processing Group ------------------------------------------------------------------------------- From: Davide Anguita > Are you using PVM today? Yes, version 3. > > What are your application(s)? Neural nets (back-prop) on workstation clusters. > > How many and what kind of machines have you used with PVM? > We are currently running it on a cluster of three ws: IBM 6000/550 HP 9000/720 HP 9000/750 and HP 9000/735 soon (hopefully). Thanks for developing such a good tool ! ========================================================================== Davide Anguita DIBE Office: +39-10-3532192 University of Genova Via all'Opera Pia 11a Fax: +39-10-3532175 16145 Genova e-mail: anguita@dibe.unige.it ITALY ========================================================================== ------------------------------------------------------------------------------- From: sarwar@up.edu (Mansoor Sarwar) Q.1. Yes, I am using PVM today. Q.2. I am just a beginner. My goal is to use PVM as a tool to teach distributed computing. Q.2. 25 Suns workstations and a 12-processor Sequent Symmetry. Request ------- I attended "Concurrent Computing with PVM" at HPDC-2. Dr. Vaidy Sunderam, the presenter, said that we could get the sources for the examples discussed in the tutorial by sending him an email. I have sent him two messages already but haven't heard from him yet. Maybe he is out of town. Could you please send me sources for Matrix Multiply, Sorting , 15-puzzle problem, Mandelbrot Set Computation, and some other examples? I would very much appreciate this. Regards, Mansoor ------------------------------------------------------------------------------- From: ftacnikj@fnmwx4.fnal.gov (Jan Ftacnik) Dear PVM Projec Team, I am very happy about you making a survey about the usage of PVM system. Question: Are you using PVM today? Answer: Yes, fairly actively. I got into using PVM only after the article about PVM in Computers in Physics, but now I am rather heavy user. Question: What are your application(s)? Answer: I am a high energy physicist and I happen to use PVM now for paralelizing mostly old Monte Carlo codes of various kinds. It is mostly fairly easy, because it involves very little communication, all processors are generating events independently, but still the resulting effect of running Monte Carlo generator on 30 machines at once to compare with one is impressive. I intend to use PVM for solving coupled differential equations and for lattice gauge calculations in the near future. Question: How many and what kind of machines have you used with PVM? Answer: I have installed and used PVM on 3 different kind of machines. First was a cluster of 8 Silicon Graphics Indigos, than pair of RS6000 and finaly around 30 SUN worksations of various types - and yes I have also run in mixed machines environment (mostly for testing, not real production runs). Suggestion: My only complaint about PVM (version 3.1) is, that the scheme of detection if pvm deamon is already running on the given machine for the given user is not flexible enough. I have run into problems both on SGI and on SUN and I have to use my patch in order to be able to use PVM at all. Problem is fairly simple: If the setup of the workstation cluster is done in such a way that all machines have the same /tmp directory, the first deamon to be started writes his file into that area, but very frequently second procesor has for the same user the same uid number and is also trying to write into the same /tmp area. But first it checks whether such file is already there. It sure is, but because of circumstances not written by the same deamon. So you cannot normaly start in this setup more than one pvm deamon - so effectively you cannot use the whole cluster. I am just now ignoring the existence of the file that the deamon has to write into /tmp directory, but it is a kludge. I didn't want to introduce new scheme into my source code, so for the time beening I am only effectively avoiding to check for the old file in /tmp area. The solution to this problem is fairly simple - one can produce more complicated name instead of /tmp/pvml.uid one could think about /tmp/pvml.machineid.uid or something along these lines. One minor suggestion: PVM is now fairly strongly UNIX-flavored because of the usage of "fork" philosophy. With the recent port to VAX VMS and possible other ports (I for myself am trying to port PVM for OS2 system with TCP/IP add ons, I can think about possible WIN NT port in the future) I would suggest to change those few "forks" to "exec" philosophy even at the expense of little more coding but for the sake of easier porting to other (understand non-UNIX platforms). Thank upou for the great job you are doing with the development of PVM and also for the interest in its usage. Keep it up ! Jan Ftacnik ftacnikj@fnal05.fnal.gov ------------------------------------------------------------------------------- From: pancake@chert.CS.ORST.EDU (Cherri Pancake) I am using PVM to develop a hands-on, self-guided tutorial that will be available for online distribution by November. It includes text, sample code, and "helpful" tools. We have PVM mounted on: HP cluster IBM RS/6000 cluster Sun cluster and are currently (trying to) mounting it on an iPSC/860. Could I please have a copy of the results of your survey? Cherri Pancake Dept. of Computer Science Oregon State University Corvallis, OR 97331 503-737-2109 (fax) 503-737-3014 pancake@cs.orst.edu ------------------------------------------------------------------------------- From: rsc8924@cfdd41.boeing.com (Dick Coffey) >Are you using PVM today? Experimentally. >What are your application(s)? No real ones yet. >How many and what kind of machines have you used with PVM? Up to six SGIs. Dick Coffey Boeing CFD Lab rsc8924@cfdd41.boeing.com 206-865-6160 ------------------------------------------------------------------------------- From: simonson@zinfandel.u-strasbg.fr (Thomas Simonson) I'm using pvm currently. My application is the molecular dynamics package Xplor (A. Brunger, Yale) I've used HP 730's and SGI Indigo's. Tom Simonson ------------------------------------------------------------------------------- From: GELFIPRO@icil64.cilea.it I have only recently started to use PVM. So far, I have used it with only three IBM Risc/6000. I intend to use PVM to develop a Non-linera Finite Element Progra m So far I have only done some small test and I had no real problem. Best regards, and keep up your excellent work! Paolo Riva ------------------------------------------------------------------------------- From: Tomohiro Kudoh > Are you using PVM today? Yes. > What are your application(s)? We are implementing a language. The language is called NCC and is specially designed for data-parallel scientific applications. > How many and what kind of machines have you used with PVM? Currently we are working on two Sparc stations. Since peoples in Kyushu-University are working on implementing PVM on a AP-1000 (Fujitsu's parallel machine with 256 processors), we plan to use it as soon as it becomes available. > Any comments and suggestions for improvement are welcome. Are there any patch to port PVM to Solaris 2.X environment? We are very pleased with PVM environment. We really thank you for your efforts to provide PVM. -------------------------------------------------------------- Tomohiro Kudoh Dept. of Information Technology Tokyo Engineering University kudoh@cntr.teu.ac.jp kudoh@aa.cs.keio.ac.jp ------------------------------------------------------------------------------- From: Viswanath Gopalakrishnan Applications: Performing tabulations on US Census Data. Data Sets are currently around 500Mbytes right now. Machines on which PVM was used: 1) A cluster of 8 HP900/735s on a FDDI token ring. 2) A 16 node SP/1. Suggestions: Being at an intermediate level user (3 months) as far as PVM goes, I have the following suggestions to offer : a) The user guide is good, but a couple of examples showing what kinds of solutions can be / cannot be implemented using PVM would be of great help. This would help people who are not very computer literate in terms of message-passing vs. shared memory decide for themselves whether PVM is the way to go for their specific solution. b) A debugger is a must with PVM, and should be made available with PVM as such. c) A list of the most common error messages and their likely causes would be a welcome addition to the user guide. This would be very helpful for the beginners. A list of "gotchas" for PVM users would also be great. d) A lot of users must have developed stuff to handle some routine tasks for PVM applications. eg. the pvmcleanup script(s) developed by Blaise Barney of the Cornell Theory Center. This could be collected and put in some kind of a contrib directory within the PVM distribution. Thanks Vishy =========================================================================== Viswanath "Vishy" Gopalakrishnan, vishy@mail.ciesin.org Consortium for International Earth Science Information Network [CIESIN] =========================================================================== ------------------------------------------------------------------------------- From: FSG/GMN-001 Dear PVM TEAM, I did not use PVM yet since my application was not ready for production, and then I broke my arm. I plan to use PVM on a network of 6 HP workstations as soon as the workstations are installed. My application consists in training neural networks on a number of processes. With PVM, I expect to use one script to run a set of training runs using all available computers. I will send you a more specific description when the software is running. Thanks again for making PVM available Richard Lemieux lemieux@gmn.ulaval.ca ------------------------------------------------------------------------------- From: scpa041@mailer.cira.it (Schiano P.) Dear PVM project Team, regarding your E-mail of August, 16 following there are the answer to your questions: 1) Yes, we are currently using PVM (2.4 version) 2) CIRA is the Italian Aerospace Research Center so the largest part of PVM based applications are CFD codes. 3) We currently have installed PVM on our cluster of IBM workstations and on the CONVEX Metaseries (8 HP-PA 735 processors), but we also used it for distributed applications (a CFD solver running on CRAY and an interactive particle tracer running on a SGI/INDIGO). Best regards Pasquale Schiano CIRA Via Maiorise Capua(CE) ITALY e-mail scpa041@mailer.cira.it ------------------------------------------------------------------------------ From: rz2a049@rzaixsrv2.rrz.uni-hamburg.de (Thomas Kochanek) Hi, here our answers to your survey questions: > Are you using PVM today? Yes indeed, we are. > What are your application(s)? Right now we are using PVM for mathematical problems like: o Transient Analysis of Electrical Circuits o Multirate Integration > How many and what kind of machines have you used with PVM? Our main machines used here are IBM RS/6000 workstations. Most of them 220s and 320s and a cluster of 4 550s. We also used it on a Sun 10 and installed it on a NeXT and I486, but only for experimental purposes. Those machines do not have enough computing power for our applications. We ported it to a Siemens/Fujitsu S100 vector computer but only for version 2.4.1. That's all for now but I hope we will have more users of PVM in the future. Some weeks ago we taught a class about version 3 and at the end of this month there will be an article in our magazine about how to use PVM. Bye, Thomas Kochanek ------------------------------------------------------------------------------- From: david@informatics.rutherford.ac.uk >Are you using PVM today? Yes. >What are your application(s)? I investigate PVM more that I use it in a production sense, as I work in a Parallel Processing group. Applications include ray-tracing, Gaussian Elimination and Hydrodynamic code. We have used PVM to emulate a hypercube network on a network of SUN4s. >How many and what kind of machines have you used with PVM? Sun 3/Sun 4/RS6000 The biggest set-up was about 16 Sun 4s. We use PVM for teaching, on a network of 10 Sun 4s. >Any comments and suggestions for improvement are welcome. A tidier programming interface based on a well defined machine model! I would like to see more ports to truely parallel machines e.g. networks of transputers (T800s and T9000s). Thanks for the product!, David Johnston (david@uk.ac.rl,inf) P.S. We will be attempting a port to Windows NT shortly. Any contacts here? ------------------------------------------------------------------------------- From: pvm@estyp13.estec.esa.nl We are currently using PVM on a modest number of machines: - a cluster of 8 IBM 580's - 2 shared memory systems (SGI 380 and SGI Challenge) - a cluster of the 2 8 processoer SGI machines and 6 SGI Indigos We are using it for the spawining of the processors and the communication between blocks in our multi-domain Navier-Stokes solvers. Martin Spel, European Space Agency. ------------------------------------------------------------------------------- From: Sanjay Raina Hi, Here is our response to your questions: > Are you using PVM today? Yes. > What are your application(s)? Various (from customers). > How many and what kind of machines have you used with PVM? An optimised version of PVM for the CS-2 mpp. ~Sanjay ------ Sanjay Raina | Phone : +44 454 616171 Meiko Limited | FAX : +44 454 618188 650 Aztec West | E-Mail: sanjay@meiko.co.uk Bristol BS12 4SD | ------------------------------------------------------------------------------- From: DWILLIAM@stetson.bitnet I am using PVM in an undergraduate class I am teaching on parallel programming during the fall 93 term. We have a network of 9 sun sparcs to run it on. Donna J. Williams Assoc professor Stetson Univ. Box 8330 DeLand, FL 32720 ------------------------------------------------------------------------------- From: Melancon Charles > > Are you using PVM today? It is under evaluation for future use with our queuing systems (UTOPIA). > > What are your application(s)? Finite element programs and Fluid analysies programs > > How many and what kind of machines have you used with PVM? 20 x HP-720 (64Mb, 1.2Gb) running HP-UX 9.01 16 x HP-730 (64Mb, 2Gb) running HP-UX 9.01 3 x HP-755 (384 MB, 10GB) running HP-UX 9.01 2 x IBM 580 (256 Mb, 10Gb) running AIX 2.3 > > Any comments and suggestions for improvement are welcome. No Comments yet ! It look very nice. Charles Melancon (Chuck) Numerical Methods B, 01SA4 Pratt and Whitney Canada Inc. 1000 Marie-Victorin Longueuil, Quebec, Canada, J4G 1A1 Tel : (514) 677-9411 x-4246 Technet : 8-488-4246 Internet: es30392@lilac.pwc.utc.com :-) ------------------------------------------------------------------------------- From: paw@cfm.brown.edu (Patrick Worfolk) Are you using PVM today? YES - we are developing software using PVM and running it. What are your application(s)? We work in the area of dynamical systems. The typical problem is to solve initial value problems for a large number of initial conditions. The idea is to picture the geometry of a flow or to scan a section of parameter space. We are using PVM in conjunction with the software package DsTool developed at Cornell's Center for Applied Mathematics (by myself and others). How many and what kind of machines have you used with PVM? We use PVM on Sun SPARC stations, on RS6000s, and on the new IBM SP1. We have used anywhere from 4-16 cpus. - Patrick. ========================= Patrick A. Worfolk Division of Applied Math Brown University 182 George St. Providence, RI 02912 401/863-2463 paw@cfm.brown.edu ======================== ------------------------------------------------------------------------------- From: Steve Thomas > Are you using PVM today? Yes > > What are your application(s)? Atmospheric and oceanic modeling > > How many and what kind of machines have you used with PVM? Maximum 20 machines in a configuration so far. SGI (IRIS, Indigo, Power Series), MIPS 4680 (EP/IX OS) HP 9000, SUN4. > > Any comments and suggestions for improvement are welcome. > Please send the information to pvm-survey@cs.utk.edu. > Your help is much appreciated. > Keep up the good work ! A very professional job. -- -------------------------------------------------------------------- Steve Thomas Recherche en prevision numerique sthomas@rpn.aes.doe.ca Environnement Canada tel: (514) 421-4769 2121, Route Trans-Canadienne fax: (514) 421-2106 Dorval, Quebec H9P 1J3 -------------------------------------------------------------------- ------------------------------------------------------------------------------- From: jeb@altair.com (Jim E. Brancheau) Currently my use of PVM is a part time effort (hobby). I have developed a Finite Element program that is currently running on SUN and SGI machines. I will soon port to CRAY. I'm learning fast that the communication time of standard ethernet is a major bottleneck to easily achieving my objectives. Although the timing has not been favorable yet PVM has allowed large problems to be solved using a number of relatively small workstations. My current solver uses a sparce iterative technique. Your software has been stable and useful. Annoucement of new versions by E-mail would be nice. I was using version 2.4 and just found out by chance about 3.0. Thanks for your efforts Jim Brancheau Altair Engineering ------------------------------------------------------------------------------- From: Giovanni Cignoni > > Are you using PVM today? Our attempt of using PVM in the design of oiXos failed because of a PVM constraint: processes belonging to the PVM machine must have the same user ID. This is in contrast with the oiXos requirement to support many different users working together. > What are your application(s)? We attempted to use PVM as a network communication layer for oiXos, the distributed runtime support of Oikos. Oikos is an on-going project at the Department of Computer Science of the University of Pisa, under direction of Prof. Carlo Montangero. Aim of project Oikos is the specification and the enactment of the software process. > How many and what kind of machines have you used with PVM? We have installed PVM 3.0 and experimented with it using 9 Sun 4 and one HP 9000 belonging to the Department network. By the way, we were positively impressed by the high quality of your product and able to perform some significant experiments in a short time. PVM still is a good candidate for our runtime support system provided that the multi-user problem can be solved. Best wishes, Giovanni A. Cignoni ------------------------------------------------------------------------------- From: philippe@dosadi.archipel.fr (Jean-Laurent Philippe) >> >> Are you using PVM today? YES, we are using PVM today. >> >> What are your application(s)? Archipel does not develop real applications in the PVM environment, only programs for demos and tests, for . Our work with PVM consists in porting PVM on a distributed-memory machine based on Tranputers. The OS is based on Chorus/MiX (parallel UNIX). Part of the work has been presented at PVMUG last May, in a paper entitled "PVM 3.0 on VOLVOX machines" written by Panchaud, Philippe and Tricot. >> >> How many and what kind of machines have you used with PVM? Archipel VOLVOX machines use the Chorus/MiX OS. Each node is a Transputer, and we are working on letting PVM use the i860 processor that is availble to each Transputer. Each Transputer is able to run several PVM tasks (time slicing). At the present time, Chorus/MiX is under development and it is limited to 12 Transputers. Our tests have been performed with 8 transputers with one or more PVM tasks on each. The most important aspect of PVM on VOLVOX machines is that the PVM source code of the applications written by the user do not have to include any source code targeted to the VOLVOX machine. ANY PVM APPLICATION RUNS ON THE VOLVOX MACHINES WITHOUT ANY CHANGES IN THE SOURCE OF THE APPLICATION. PORTABILITY IS TOTAL. >> For any request about PVM on VOLVOX MACHINES, please contact Jean-Laurent PHILIPPE ARCHIPEL philippe@archipel.fr ------------------------------------------------------------------------------- From: Ann Petrie > Are you using PVM today? Yes > What are your application(s)? As part of my work in the Computing Service, I am trying to get the feel of PVM on a colleague's Super Lattice program and comparing how the problem is programmed in Linda and pvm > How many and what kind of machines have you used with PVM? SUN4 (Sparc IPC) HPPA HP 710/730 > Any comments and suggestions for improvement are welcome. I was pleased by the quick response I got to the query that I sent about problems building hence (to hence@msr.epm.ornl.gov) Ann Petrie ------------------------------------------------------------------------------- From: pramanic@vnet.IBM.COM To: pvm-survey@cs.utk.edu Subject: pvm-survey Are you using PVM today? YES What are your application(s)? distributed timing How many and what kind of machines have you used with PVM? RS6000s (upto 10) ------------------------------------------------------------------------------- From: dave@sarah.lerc.nasa.gov (David A. Thompson) 1) Are you using PVM today? I have used it in the past and suspect to be using it in the near future. 2) What are your application(s)? I was writing a simple program to farm out jobs when machines became idle (basically run a single batch job on each machine). i had also put together a simple program to get approximate performance of various machines and was going to try to determine which machine was to be given the longer running cases. The aplication in the future is to take a time dependent compressible fluid code (once the code works on a single processor) and modify it to use a graphics machine for display of data (Probably SGI) and a assortment of other machines to do the computational work (probably IBM/R6000's) The code currently only saves data for a 2 d segment of the solution and the resulting file is approx 300 MB the 3d data file would be approx 36 times this size. So the initial use will be to elimanate the need for the data file by processing the graphics immediatly. I also suspect there to be other programs that i use pvm on but i don't know what they wil be yet. 3) How many and what kind of machines have you used with PVM? I have probably had about 64+ machines configured into the pvm package at one time. The types are SUN3, SUN4, SGI, IBM/RS6000. (Eventually on the HyperCube, CRAY and others) Sorry for the delay in replying but i was very busy the last couple of weeks +-----------------------------------------------------------------------------+ | David Thompson (216) 972-6776 | | The University of Akron Bitnet: r1dat@akronvm | | Mathematical Sciences Internet: dave@zippysun.math.uakron.edu | | Akron, Ohio dave@sarah.lerc.nasa.gov | +-----------------------------------------------------------------------------+ ------------------------------------------------------------------------------- From: krzys@slc1.iinf.gliwice.edu.pl (Krzysztof Pierzchala) > > Are you using PVM today? Yes. > What are your application(s)? We haven't got any our application yet. > How many and what kind of machines have you used with PVM? Now we using 1 SPARCstation IPX and 2 SPARCstation SLC, but we are going to buy some SUN Classic and LX and one SPARCserver 10. Best wishes, Krzysztof Pierzchala ------------------------------------------------------------------------------- From: Andrea.Holtz@gmd.de (Andrea Holtz) > >Are you using PVM today? yes > >What are your application(s)? We're programming with pvm. Our development is a general parallel runtime environment for applications running on several different machines. > >How many and what kind of machines have you used with PVM? Now in the phase of development we are mainly using SUNs. The applications for which we are developing the runtime environment run on cm2, cm5 and Parix. We are still waiting for the PVM-Installation on Parix, which should be released soon. Andrea Holtz ------------------------------------------------------------------------------- From: mauricio@fee.unicamp.br (Mauricio Ferreira Magalhaes) > Are you using PVM today? Yes. > What are your applications? I'm using PVM in a laboratory for undegraduate students (Computer Engineering). > What are your machines? I'm giving de laboratory in a plataform based on 6 RISC/6000 (Model 320) and 1 server (Model 520) and 2 Xterminal, connected by a token ring of 16 Mbits. ------------------------------------------------------------------------------- From: vincent@pastek.cray.com (Vincent LE BARAZER) >>> >>> Are you using PVM today? Yes. >>> >>> What are your application(s)? Porting of multi-processors single-machine applications. >>> >>> How many and what kind of machines have you used with PVM? Sun4, SGI and CRAY YMP & T3D . ------------------------------------------------------------------------------- From: detar@hetheory.physics.utah.edu (Carleton DeTar) >Are you using PVM today? Yes >What are your application(s)? Lattice Gauge Theory >How many and what kind of machines have you used with PVM? Only RS6000 cluster and SP1 Thanks. I may want to send comments after I have 3.1 up and running. ------------------------------------------------------------------------------- From: admin@ibmserv.edu.tw Dear PVM Team, Yes, I am using PVM for various studies of parallel programming on RS/6000, I find PVM lacks of collective communication functions, such as concat, combine, and many others. In addition, the manual for PVM V3 is a draft one, any final draft? regards Cinloon ------------------------------------------------------------------------------- From: detar@hetheory.physics.utah.edu (Carleton DeTar) It turns out I got PVM 3.1 running sooner than I thought, so here is an addendum to my message of yesterday. >Any comments and suggestions for improvement are welcome. >Please send the information to pvm-survey@cs.utk.edu. >Your help is much appreciated. My application requires inverting a large, sparse, homogeneous matrix on a 4-dimensional regular grid. The code is SPMD with the grid distributed uniformly over the compute nodes. A problem of this type tends to have a rather high communication to computation requirement as the granularity is reduced, since the surface to volume ratio grows rapidly with decreasing volume. At the moment, I find that the optimum minimum granularity with PVM puts 4^4 grid points on a compute node. I would like to cut this by at least an order of magnitude in order to make progress with the computation and make an RS600 cluster comparable to an Intel iPSC/860 for my application. At the moment I am apparently held up by large software latencies inherent in PVM, which eat up about 20% of the available cycles at this granularity. To try to minimize this latency, I pass messages of raw bytes on a homogeneous architecture (RS/6000's on a cluster). In my case, after setting up the computational environment, the only pvm calls need be interprocess message passing. If there were some way to get even closer to the wire, that would be very helpful, perhaps even bypassing the pvmd's after things are set up and going straight to the processes themselves. Best regards, Carleton DeTar ------------------------------------------------------------------------------- From: Thilo Kielmann > > Are you using PVM today? yes > > What are your application(s)? We experiment with a parallelizing compiler. (research in computer science) > > How many and what kind of machines have you used with PVM? A handful SUN 4 workstations. Best wishes -- Thilo Kielmann University of Darmstadt Dept. of Computer Science (FB 20) /\,/\ Magdalenenstr. 11c / u u \ D-64289 Darmstadt \<_Y_>/ Germany / " \ ------------------------------------------------------------------------------- From: pete@samson.idec.sdl.usu.edu (Peter Krull) I am currently using Express to handle message passing for an IR image data analysis application. I had run into some limitations with Express and wanted to compare it to PVM. Deadlines have overruled the analysis, however, and I have yet to do any meaningful comparisons. -- pete@samson.idec.sdl.usu.edu Peter Krull Space Dynamics Lab 1747 Research Parkway RP2 Logan, UT 84321 (801) 750-4857 ------------------------------------------------------------------------------- From: Mario Daberkow Hello PVM! I am working in the field of constructive algebraic number theory here at the TU-BERLINi under Prof. Pohst. Our research group develops the computer programm KANT for computations in algebraic number fields. So far we are using PVM for the computation of / for -- maximal orders -- principal ideal tests We plan to implent a few more alogorithms under PVM. The speedup is quite impressive (faktor up to 10). Mario Daberkow (daberkow@math.tu-berlin.de) ------------------------------------------------------------------------------- From: Luc.Giraud@cerfacs.fr Hi, Soory for the delay before replying, > Are you using PVM today? Yes we still use PVM version 2.4, because we cannot run any example with PVM 3.1 on our IBM/RS6000 network. > What are your application(s)? Our applications are mainly in CFD, linear algebra and for the coupling of different model for example a work has been done for coupling numerical models of the ocean and atmosphere. > How many and what kind of machines have you used with PVM? We use PVM on our network of workstations IBM and SUN, on a BBN TC2000 (where the PVM source has been modified to take advantage of the virtual shared memory) and on a CRAY-2 for the coupling example. Best Regards, Luc ------------------------------------------------------------------------------- From: "Dr. Uwe Feldmann" Dr. U. Wever Siemens AG Corporate Research and Development Dept. ZFE BT SE 43 D-81730 Munich 1. Since one year we are working with PVM. Before we have tested a lot of parallel platforms. PVM turned out to be the best choice for scientific computing. 2. In our department we have developed a lot of parallel software: A parallel circuit simulator, a seismic simulator and a dopant diffusion simulator. The underlying parallelization techniques are mainly domain decomposition methods. But also direct solver parallelization are used. 3. PVM seems to be a very stable and reliable parallel platform. Also the performance is very good. Problems arises with node-IO. If one node fails the IO-buffer will not be emptied. ------------------------------------------------------------------------------- From: bjam@ctrss1.lanl.gov (Bertrand J. Meltz) 1_ Yes, I use PVM. 2_ My applications : - Climate modeling - Low speed flow (incompressible fluid dynamics) - More generally, any form of finite difference CFD 3_ In chronological order: - Sun Workstations - IBM cluster - Cray T3D emulator hopefully soon : - CM5 - Intel IPSC and Paragon - any other MPP I can have access to. Comments : It's great! Works fine! A few complaints : some functions (like pvmfpack/unpack) are illegal in Fortran, where you can't call routines with arguments of varying type. Most compilers don't object, but some do (like the F90 by NAG). Also, some argument values for these functions are missing, like INTEGER8 (some machines DO have integers 8-bytes long), but this may already have been corrected. Bertrand Meltz Group T-3, MS K723 Los Alamos National Laboratory Los Alamos, NM 87545 Tel : (505) 667 0954 Fax : (505) 665 5926 e_mail : bjam@lanl.gov ------------------------------------------------------------------------------- From: Jamie BULBECK In answer to your questions: No, I am not using PVM at the moment. I intend using PVM rather a lot in the near future, as I am conducting postgraduate research into direct numerical simulation of turbulent fluid flows, and PVM will come in very handy for developing simulation code before porting it to MIMD parallel supercomputers. Currently I've had PVM up and running on an Silicon Graphics 4D 25 and two Suns. I couldn't get it to compile on a DEC Station 3100. I've also used it on a cluster of IBM RS/6000s. Regards Jaime Bulbeck Mechanical and Manufacturing Engineering University of Melbourne Australia ------------------------------------------------------------------------------- From: Peter Corbett 1) Yes, I plan to use PVM today. 2) My application is a three-dimensional compressible Navier-Stokes solver. 3) The von Karman Institute has a "farm" of four Digital AXP platforms: the primary goal is to use these in parallel with an FDDI hookup. However, a heterogenous machine was set up composed of AXPs and PMAXs which worked just fine, although the PMAXs run about an order of magnitude slower than the AXPs... Insofar as comments: I was pleasantly surprised by how easy it was to paralle- lise the algorithm. However, the docs, while definitely a good start, require some polishing...and in some areas are downright patchy. Additionally, at the moment I may be having some troubles with buffer sizes... Sincerely, Peter Corbett ------------------------------------------------------------------------------- From: Eric Grosse 908-582-5828 > > Are you using PVM today? Yes. > What are your application(s)? domain decomposition solution of systems of partial differential equations. > How many and what kind of machines have you used with PVM? Small number (less than 7) of R4000 Indigo on FDDI network. I find that coding my application directly with sockets gives me about a factor of 3 speedup over pvm. I did turn on pvm_advise(PvmRouteDirect), which did help; the "cost" of using pvm used to be a factor of 6. Best wishes, Eric ------------------------------------------------------------------------------- From: deflorio@fourier.rob.csata.it Hi, > Could you take a moment to answer three questions. it's a pleasure for me to answer to your questions. Q: Are you using PVM today? A: Yes Q: What are your application(s)? A: We are developing two applications: the first one is called PvmLinda and implements the linda model on top of PVM, allowing people to use a mixture of the message passing and of the live data structure computational models; the second one is called RHL (Remote Host Library) and is a four-function library for executing remote functions without any care for communication technicalities. Any question is welcome. Q: How many and what kind of machines have you used with PVM? A: 4 RS6K, 1 SUN4, 1 SUN3, 2 HP9K, 1 CNVX Thanks for all your efforts, - Vincenzo De Florio deflorio@fourier.rob.csata.it ------------------------------------------------------------------------------- From: lance c burton > >Are you using PVM today? > not necessarily today, but as a part of my ongoing research >What are your application(s)? > hetergeneous computing with Object-Oriented Fortran >How many and what kind of machines have you used with PVM? > SUN4 SGI IPSC -- . - . . | . . | . lance c burton ( - - + - - ) burtonl@erc.msstate.edu . | . . | . - ------------------------------------------------------------------------------- From: rpoll@bsys004.fob.ford.com The following applies to the use of PVM in Ford in Europe for Computer Aided Engineering. I don't know of any other users in Europe, but there certainly are people in Ford in the US doing some work. >Are you using PVM today? We are not actively using PVM here currently. The copy we have is being used for an informal 'evaluation' of the system and it's capabilities. >What are your application(s)? Our applications WOULD be primarily FE codes like NASTRAN, STAR-CD, PAMCRASH RADIOSS etc. Most of our applications are third party, but we do develop some software in-house. >How many and what kind of machines have you used with PVM? The 'evaluation' to date has been using DEC MIPS hardware, up to about 30 workstations. My general comments on PVM are: -The main strengths of PVM are it's wide platform support and it's widespread use. -I believe that PVM will promote the porting of third party applications (which is our main area of interest) to parallel environments. -PVM is not yet 'industrial strength'. It needs to be able to handle nodes in the 'virtual machine' going down, or running out of disk etc. in a more robust way. Also front end tools to 'manage' the machine (queuing, load balancing and sheduling) are required. -I believe that PVM will be most successful running on a dedicated cluster of machines. Use of 'free cycles' from User's workstations would be very difficult to manage in an industrial environment. -I believe that PVM is a stepping stone to virtual shared memory machines, because the programming effort will be lower, and also because there is potential for less software to be involved in moving data, and hence lower latencies. How long it will take for virtual shared to become mature is a different question though! I would be very interested to hear the results of your survey, and to hear other User's comments - do you plan to publish anything? Robert Poll Ford Motor Company Ltd. Dunton, England Mail: rpoll@ecs01.pms.ford.com ------------------------------------------------------------------------------- From: Mark Kessler Are you using PVM today? Intermittently. We have done some preliminary (simple) applications for a proof of concept type evaluation, but not gone beyond that stage. What are your application(s)? We are interested in data compression (video and still image), statistical modeling for circuit simulations, FEM and Partial Diff. Eq. type applications. How many and what kind of machines have you used with PVM? We currently have one principle investigator (me) using the software on a collection of Sun's and Decstations (primarily Sun). Any comments and suggestions for improvement are welcome. The HeNCE software has proven to be extremely useful to create PVM applications. We immediately wrote a little C interface to invoke any UNIX shell level command and this allowed us to provide users with a tool to design and run thier own course grained (IE the corresponding shell level commands, such as a command to invoke a spice like circuit simulation) parrallel applications. By and large I was favorably impressed however.... A more robust decision making schema that could be used for arguments to wrappers would be very useful. For example, the ability to call a routine that changes an Input/Output parameter and then use the resulting value to make decisions in my looping structures or fanout structures would be useful in our design optimization and statistical modeling applications. An immediate use would be to allow us to drive these applications by tabuluar input files created outside of the PVM domain. A run time only HeNCE tool would be very useful. I (generically speaking) could create a collection of tools that others who were not familiar with the PVM system or the HeNCE compilation procedures could use to implement parallel solutions within an applicaiton specific domain. The user interface was quite good for Public Domain Software, but it still wasn't quite industrial strength so to speak. I want to convey that my impression of PVM (particularly with a front end tool like HeNCE) was very positive and that the assistance and information I recieved was appropriate to level of help required. I trust these comments are taken in the positive spirit in which they are given. Best Regards Mark Kessler ------------------------------------------------------------------------------- From: fred@wubcl.wustl.EDU (message body lost by mailer) Fred U. Rosenberger fred@wubcl.wustl.EDU Biomedical Computer Laboratory Voice: (314) 362-3124 Washington University Fax (e-mail preferred): (314) 362-0234 700 S. Euclid Ave. St. Louis, MO 63110 -------------------------------------------------------------------------------