PSTSWM SP3-200 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-200 Protocol Performance

(distributed FFT experiment A3)

Date/Person: June 6, 1999 / P. Worley
Platform: IBM SP3 at Oak Ridge National Laboratory (morgan.ccs.ornl.gov):
     62 2-way Winterhawk I SMP nodes (200 MHz POWER3 with 4MB L2 cache, equivalent to RS/6000 Model 260)
Environment: AIX 4.3.2;   PSSP 3.1
Code Version: 6.5
Compilation Options: mpxlf -O3 -qarch=auto -qtune=auto -qcache=auto
Link Options: -bmaxdata:0x70000000
Math Library: ESSL
Communication Library: MPI over the switch using user space for internode communication
MPI using IP for intranode communication
Parallel Algorithm: distributed FFT
Number of Timesteps: 12
Results:

16x1 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.9443e-02   0.39   0.29   0.90 
Three Fastest
Protocols
1st2nd3rd
  c2   c3   d2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   1   13 

32x1 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.6003e-02   0.35   0.36   0.65 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   a5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   4   9 

8x1 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.3340e-01   0.20   0.19   0.70 
Three Fastest
Protocols
1st2nd3rd
  c5   a5   c4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   6   16 

16x1 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.6860e-01   0.20   0.21   0.78 
Three Fastest
Protocols
1st2nd3rd
  a4   a5   a1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   5   20 

32x1 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.5521e-01   0.17   0.17   0.57 
Three Fastest
Protocols
1st2nd3rd
  c5   c4   d4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   4   24 

8x1 Processors / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.9613e+00   0.14   0.13   0.73 
Three Fastest
Protocols
1st2nd3rd
  d5   c5   a1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   12   23 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:16:09 EDT.