PSTSWM SP3-375 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-375 Winterhawk II Protocol Performance

(transpose LT experiment I-C / O(P) sendrecv transpose algorithm)

Date/Person: December 12, 1999 / P. Worley
Platform: IBM SP3 at Oak Ridge National Laboratory (bobcat.ccs.ornl.gov):
     62 4-way Winterhawk II SMP nodes (375 MHz POWER3 with 8MB L2 cache)
Environment: AIX 4.3.3;   PSSP 3.1.1
Code Version: 6.6.4
Compilation Options: mpxlf -O3 -qarch=auto -qtune=auto -qcache=auto
Link Options: -bmaxdata:0x70000000
Environment Variables: export MP_SHARED_MEMORY="yes"
Math Library: ESSL
Communication Library: MPI over the switch using user space for internode communication
MPI using shared memory for intranode communication
Parallel Algorithm: srtrans
(assuming distributed FFT)
Number of Timesteps: 12
Results:

4x16 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.2879e-01   0.21   0.19   0.44 
Three Fastest
Protocols
1st2nd3rd
  e3   e4   e5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   3   20 

2x32 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.8106e-01   0.37   0.37   0.76 
Three Fastest
Protocols
1st2nd3rd
  e3   e4   e5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   8 

8x8 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.6336e-01   0.05   0.05   0.12 
Three Fastest
Protocols
1st2nd3rd
  e3   e5   e4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   16   31 

4x16 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.5262e-01   0.11   0.10   0.24 
Three Fastest
Protocols
1st2nd3rd
  e3   e4   c4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   8   31 

2x32 Processors (Column Major Ordering) / Problem T42L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  8.2541e-01   0.15   0.12   0.37 
Three Fastest
Protocols
1st2nd3rd
  e3   c4   e4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   5   26 

8x8 Processors (Column Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.3069e+00   0.05   0.04   0.13 
Three Fastest
Protocols
1st2nd3rd
  e2   a4   c1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   20   31 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:19:39 EDT.