PSTSWM Paragon Communication Protocol Performance

Performance Studies using

PSTSWM

Intel Paragon Protocol Performance

(distributed LT experiment A1 / ring 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: ringsum
Partition: 1x8, 1x16, or 1x32
Results:

1x16 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.3275e-01   0.19   0.15   0.50 
Three Fastest
Protocols
1st2nd3rd
  c2   c4   a2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   16 

1x32 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.8990e-01   0.25   0.22   0.64 
Three Fastest
Protocols
1st2nd3rd
  c2   c4   a3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   15 

1x8 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.2053e+00   0.03   0.02   0.22 
Three Fastest
Protocols
1st2nd3rd
  c2   c4   c5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   24   25 

1x16 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.0365e+00   0.04   0.03   0.30 
Three Fastest
Protocols
1st2nd3rd
  c4   c2   c3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   20   24 

1x32 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.5960e+00   0.07   0.05   0.47 
Three Fastest
Protocols
1st2nd3rd
  c5   c3   c2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   11   24 

1x8 Processors / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.3314e+01   0.02   0.01   0.15 
Three Fastest
Protocols
1st2nd3rd
  d6   b6   b5 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  10   24   25 

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 10:29:57 EDT.