PSTSWM SP3-375 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-375 Winterhawk II Protocol Performance

(transpose FFT experiment A / 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:

16x1 Processors / Problem T10L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.0951e-01   0.34   0.35   0.71 
Three Fastest
Protocols
1st2nd3rd
  e0   e2   e1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   4   12 

32x1 Processors / Problem T10L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.0156e-01   0.38   0.38   0.79 
Three Fastest
Protocols
1st2nd3rd
  e2   e3   e0 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   4   6 

8x1 Processors / Problem T21L8
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.2936e-01   0.25   0.24   0.48 
Three Fastest
Protocols
1st2nd3rd
  e5   e4   e3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   18 

16x1 Processors / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.1633e-01   0.37   0.36   0.72 
Three Fastest
Protocols
1st2nd3rd
  e5   e4   e3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   2   9 

32x1 Processors / Problem T21L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.6067e-01   0.45   0.43   0.91 
Three Fastest
Protocols
1st2nd3rd
  e5   e4   e3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   3   5 

8x1 Processors / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  9.4285e-01   0.12   0.11   0.67 
Three Fastest
Protocols
1st2nd3rd
  e5   a4   a5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   6   29 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:19:40 EDT.