PSTSWM T3E-900 Communication Protocol Performance

Performance Studies using

PSTSWM

SGI/Cray Research T3E-900 Protocol Performance

(transpose LT experiment I-B / O(P) sendrecv transpose algorithm)

Date/Person: May 11, 1998 / P. Worley
Platform: T3E-900 at National Energy Research Scientific Computing Center (mcurie.nersc.gov):
   532 450-MHz DEC Alpha EV5 RISC processors
Environment: UNICOS/mk 2.0.2.28
mpt 1.2.0.2
Cray CF90 Version 3.0.2.1
Code Version: 6.3
Compilation Options: -dp -Oscalar3 -Wl,-Dstreams=on
Math Library: libsci
Communication Library: SHMEM
Parallel Algorithm: srtrans
(assuming distributed FFT)
Number of Timesteps: 12
Results:

4x16 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  7.8743e-02   0.07   0.04   0.45 
Three Fastest
Protocols
1st2nd3rd
  e1   a2   a1 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  5   7   11 

2x32 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  9.1199e-02   0.07   0.06   0.16 
Three Fastest
Protocols
1st2nd3rd
  c1   e1   c2 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   4   12 

8x8 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.3248e-01   0.01   0.01   0.03 
Three Fastest
Protocols
1st2nd3rd
  c1   c2   a1 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  5   12   12 

4x16 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.9655e-01   0.03   0.03   0.08 
Three Fastest
Protocols
1st2nd3rd
  c1   c2   a1 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   10   12 

2x32 Processors (Column Major Ordering) / Problem T42L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.2457e-01   0.03   0.03   0.07 
Three Fastest
Protocols
1st2nd3rd
  a2   c1   c2 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   9   12 

8x8 Processors (Column Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.2154e+00   0.01   0.01   0.02 
Three Fastest
Protocols
1st2nd3rd
  c1   a1   a2 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  6   12   12 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:27:58 EDT.