PSTSWM SP3-200 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-200 Protocol Performance

(transpose LT experiment I-B2 / O(P) swap transpose algorithm)

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: swtrans
(assuming distributed FFT)
Number of Timesteps: 12
Results:

4x16 Processors (Row Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.7743e-01   0.26   0.25   0.43 
Three Fastest
Protocols
1st2nd3rd
  e3   e4   e2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   15 

2x32 Processors (Row Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.1898e-01   0.33   0.36   0.57 
Three Fastest
Protocols
1st2nd3rd
  e5   e3   e2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   4   10 

8x8 Processors (Row Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.2354e-01   0.08   0.09   0.17 
Three Fastest
Protocols
1st2nd3rd
  e5   a4   c5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   8   31 

4x16 Processors (Row Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.5354e-01   0.15   0.14   0.33 
Three Fastest
Protocols
1st2nd3rd
  e3   e5   e4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   26 

2x32 Processors (Row Major Ordering) / Problem T42L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  6.7003e-01   0.18   0.17   0.40 
Three Fastest
Protocols
1st2nd3rd
  e5   e3   c4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   4   20 

8x8 Processors (Row Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.1257e+00   0.04   0.03   0.13 
Three Fastest
Protocols
1st2nd3rd
  a1   a4   c1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   22   31 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:17:02 EDT.