PSTSWM SP3-375 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-375 Winterhawk II Protocol Performance

(transpose FFT experiment 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
Number of Timesteps: 12
Results:

16x4 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.5225e-01   0.55   0.54   1.11 
Three Fastest
Protocols
1st2nd3rd
  e4   e5   e3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   3   5 

32x2 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.6719e-01   0.53   0.53   1.15 
Three Fastest
Protocols
1st2nd3rd
  e4   e3   e5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   5 

8x8 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.9985e-01   0.07   0.05   0.17 
Three Fastest
Protocols
1st2nd3rd
  c5   a5   c4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   13   31 

16x4 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.1893e-01   0.16   0.13   0.38 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   e4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   4   22 

32x2 Processors (Column Major Ordering) / Problem T42L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  6.2302e-01   0.22   0.16   0.56 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   a4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   3   21 

8x8 Processors (Column Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.5562e+00   0.08   0.07   0.33 
Three Fastest
Protocols
1st2nd3rd
  a5   a1   a4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   8   30 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:19:42 EDT.