PSTSWM T3D Communication Protocol Performance

Performance Studies using

PSTSWM

SGI/Cray Research T3D Protocol Performance

(transpose LT experiment II-A / O(P) sendrecv 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: srtrans
(assuming transpose FFT)
Number of Timesteps: 12
Results:

1x16 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.1790e-01   0.06   0.05   0.16 
Three Fastest
Protocols
1st2nd3rd
  e2   c2   a2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   6   12 

1x32 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.7565e-01   0.07   0.07   0.17 
Three Fastest
Protocols
1st2nd3rd
  e2   c2   a2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   5   12 

1x8 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  9.2368e-01   0.03   0.01   0.08 
Three Fastest
Protocols
1st2nd3rd
  c2   a2   e2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   10   12 

1x16 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.4735e+00   0.02   0.01   0.05 
Three Fastest
Protocols
1st2nd3rd
  c2   a2   d6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  5   10   12 

1x32 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.7938e+00   0.02   0.01   0.05 
Three Fastest
Protocols
1st2nd3rd
  c2   a2   e2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  6   11   12 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:24:07 EDT.