PSTSWM SP3-200 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-200 Protocol Performance

(distributed LT experiment B2 / ring-pipeline sum 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: ringpipe
Number of Timesteps: 12
Results:

4x16 Processors (Row Major Ordering) / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.2102e-01   0.41   0.38   0.93 
Three Fastest
Protocols
1st2nd3rd
  i5   f4   g4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   9 

2x32 Processors (Row Major Ordering) / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.2741e-01   0.76   0.71   1.63 
Three Fastest
Protocols
1st2nd3rd
  i4   i5   f5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   8 

8x8 Processors (Row Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.5358e-01   0.13   0.11   0.29 
Three Fastest
Protocols
1st2nd3rd
  f4   i4   i5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   9   46 

4x16 Processors (Row Major Ordering) / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.4312e-01   0.21   0.18   0.52 
Three Fastest
Protocols
1st2nd3rd
  f5   f4   i5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   5   35 

2x32 Processors (Row Major Ordering) / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.9711e-01   0.41   0.34   0.86 
Three Fastest
Protocols
1st2nd3rd
  i4   f4   g4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   9 

8x8 Processors (Row Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.2509e+00   0.06   0.05   0.15 
Three Fastest
Protocols
1st2nd3rd
  f3   f5   a2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  5   27   50 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:16:38 EDT.