PSTSWM T3D Communication Protocol Performance

Performance Studies using

PSTSWM

SGI/Cray Research T3D Protocol Performance

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

Date/Person: October 28, 1994 / P. Worley
Platform: T3D at Cray Research in Eagen, MN (rain):
   128 150-MHz DEC Alpha EV4 RISC processors
Environment: IS-9.0
Cray CF90 Version 0.1.2.3
Code Version: 6.1
Compilation Options: /mpp/bin/f90 -dp -Oscalar3
Math Library: none
Communication Library: SHMEM
Parallel Algorithm: swtrans
(assuming distributed FFT)
Number of Timesteps: 12
Results:

1x16 Processors / Problem T10L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.6069e-01   0.07   0.06   0.17 
Three Fastest
Protocols
1st2nd3rd
  a2   c2   d6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   6   12 

1x8 Processors / Problem T21L8
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  6.8694e-01   0.04   0.01   0.11 
Three Fastest
Protocols
1st2nd3rd
  a2   c2   d6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   8   12 

1x16 Processors / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  7.3579e-01   0.04   0.03   0.11 
Three Fastest
Protocols
1st2nd3rd
  a2   c2   d6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   8   12 

1x32 Processors / Problem T21L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  7.5312e-01   0.04   0.02   0.12 
Three Fastest
Protocols
1st2nd3rd
  a2   c2   d6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   9   12 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:24:10 EDT.