PSTSWM T3E-900 Communication Protocol Performance

Performance Studies using

PSTSWM

SGI/Cray Research T3E-900 Protocol Performance

(transpose LT experiment I-C / O(P) swap 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: MPI
Parallel Algorithm: swtrans
(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
  1.4446e-01   0.11   0.10   0.25 
Three Fastest
Protocols
1st2nd3rd
  a6   a0   a3 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   6   30 

2x32 Processors (Column Major Ordering) / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.8143e-01   0.17   0.13   0.40 
Three Fastest
Protocols
1st2nd3rd
  a6   a0   b0 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   25 

8x8 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.3517e-01   0.03   0.03   0.07 
Three Fastest
Protocols
1st2nd3rd
  c1   e0   c2 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  6   26   31 

4x16 Processors (Column Major Ordering) / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.0635e-01   0.06   0.06   0.12 
Three Fastest
Protocols
1st2nd3rd
  a6   a0   e2 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   13   31 

2x32 Processors (Column Major Ordering) / Problem T42L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  6.6538e-01   0.07   0.07   0.15 
Three Fastest
Protocols
1st2nd3rd
  a6   a0   e0 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   11   31 

8x8 Processors (Column Major Ordering) / Problem T85L32
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  3.6727e+00   0.01   0.01   0.02 
Three Fastest
Protocols
1st2nd3rd
  e0   c1   e1 
      Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  18   31   31 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:28:27 EDT.