PSTSWM Paragon Communication Protocol Performance

Performance Studies using

PSTSWM

Intel Paragon Protocol Performance

(distributed LT experiment A1 / ring-pipeline sum 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: MPI
Parallel Algorithm: ringpipe
Partition: 1x8, 1x16, or 1x32
Results:

1x16 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.5221e-01   0.29   0.29   0.81 
Three Fastest
Protocols
1st2nd3rd
  i4   i3   c2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   24 

1x32 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.6693e-01   0.29   0.26   0.90 
Three Fastest
Protocols
1st2nd3rd
  c2   i3   c1 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   5   22 

1x8 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.1028e+00   0.15   0.17   0.36 
Three Fastest
Protocols
1st2nd3rd
  i3   i4   f3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   11   49 

1x16 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.8425e+00   0.18   0.20   0.46 
Three Fastest
Protocols
1st2nd3rd
  i3   i4   i5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   5   44 

1x32 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.3540e+00   0.24   0.24   0.65 
Three Fastest
Protocols
1st2nd3rd
  i3   f3   i5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   26 

1x8 Processors / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.2954e+01   0.10   0.13   0.27 
Three Fastest
Protocols
1st2nd3rd
  f3   i3   i5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   13   49 

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 10:30:00 EDT.