PSTSWM Paragon Communication Protocol Performance

Performance Studies using

PSTSWM

Intel Paragon Protocol Performance

(transpose LT experiment II-A1 / O(P) sendrecv transpose algorithm)

Date/Person: May 13, 1998 / P. Worley
Platform: Intel Paragon XP/S 150 MP at Oak Ridge National Laboratory:
     1024 MP nodes (3 50-MHz iPSC/860 processors per node)
Environment: Paragon OSF/1 Release 1.0.4 Server 1.4 R1_4_5
f77/Paragon Paragon Version R5.0.3
Code Version: 6.3
Compilation Options: if77 -O4 -Mnodepchk -Knoieee -Msafealloc
Math Library: KAI
Communication Library: NX
Parallel Algorithm: srtrans
Partition: 1x8, 1x16, or 1x32
Number of Timesteps: 12
Results:

1x16 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.8578e-01   0.30   0.35   0.73 
Three Fastest
Protocols
1st2nd3rd
  a3   a2   a1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   5   13 

1x32 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.3784e-01   0.27   0.24   0.68 
Three Fastest
Protocols
1st2nd3rd
  a2   a3   a6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   4   15 

1x8 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.1457e+00   0.15   0.15   0.32 
Three Fastest
Protocols
1st2nd3rd
  a2   a3   a6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  6   10   22 

1x16 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.7964e+00   0.09   0.09   0.17 
Three Fastest
Protocols
1st2nd3rd
  a1   a2   a3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   9   29 

1x32 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.0499e+00   0.20   0.24   0.40 
Three Fastest
Protocols
1st2nd3rd
  a1   a6   a3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  5   8   15 

1x8 Processors / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.4441e+01   0.10   0.12   0.18 
Three Fastest
Protocols
1st2nd3rd
  b1   b0   a6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  8   9   29 

DISCUSSION

Partitions of the Paragon processor grid were used that match the processor subsets that the parallel algorithms would run on in a two dimensional data decomposition: 1x8, 1x16, and 1x32.
Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:30:12 EDT.