PSTSWM AlphaSC-500 Communication Protocol Performance

Performance Studies using

PSTSWM

Compaq AlphaServer SC Protocol Performance

(transpose LT experiment I-A1 / O(P) swap transpose algorithm)

Date/Person: November 4, 1999 / P. Worley
Platform: Compaq AlphaServer SC at Compaq (fsehpc):
     16 ES40 4-way SMP nodes (500 MHz Alpha 21264 with 4MB L2 cache)
Environment: Digital UNIX V5.0;   RMS 2.3.2
Code Version: 6.6.3
Compilation Options: f90 -O4 -assume accuracy_sensitive -math_library accurate -aling dcommons -align records -arch host -tune host
Math Library: none
Communication Library: MPI
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
  7.3792e-02   0.19   0.07   1.83 
Three Fastest
Protocols
1st2nd3rd
  e2   e3   c2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  4   11   24 

1x8 Processors / Problem T21L8
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.1814e-01   0.12   0.09   0.98 
Three Fastest
Protocols
1st2nd3rd
  e3   e2   a6 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   3   28 

1x16 Processors / Problem T21L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.7626e-01   0.13   0.07   1.54 
Three Fastest
Protocols
1st2nd3rd
  e2   a2   b2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   10   27 

1x8 Processors / Problem T42L16
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  9.6215e-01   0.04   0.04   0.12 
Three Fastest
Protocols
1st2nd3rd
  a1   a6   a3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  3   19   29 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:03:28 EDT.