PSTSWM T3E-900 Algorithm Comparison

Performance Studies using

PSTSWM

SGI/Cray Research T3E-900 Algorithm Comparison

(transpose FFT experiment C)

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
SHMEM
Number of Timesteps: 12
Results:

Transpose FFT (mpi) / Column Major Processor Ordering
Algorithm Comparison
  T21L16    T21L16    T42L16    T42L32    T42L16    T85L32 
  P=32x2     P=16x4     P=8x8     P=32x2     P=16x4     P=8x8  
  optimal algorithm   alltoallv  alltoallv  alltoallv  alltoallv  alltoallv  alltoallv 
  (alltoallv-min)/min     0.000    0.000    0.000    0.000    0.000    0.000 
  (generic-min)/min     0.312    0.234    0.046    0.091    0.085    0.006 

Transpose FFT (shmem) / Column Major Processor Ordering
Algorithm Comparison
  T21L16    T21L16    T42L16    T42L32    T42L16    T85L32 
  P=32x2     P=16x4     P=8x8     P=32x2     P=16x4     P=8x8  
  optimal algorithm   swtrans  swtrans  swtrans  srtrans  swtrans  swtrans 
  (generic-min)/min     0.208    0.134    0.032    0.052    0.058    0.021 

Transpose FFT (combined) / Column Major Processor Ordering
Communication Library Comparisons
  T21L16    T21L16    T42L16    T42L32    T42L16    T85L32 
  P=32x2     P=16x4     P=8x8     P=32x2     P=16x4     P=8x8  
  optimal library   shmem  shmem  shmem  shmem  shmem  shmem 
  (alltoallv-min)/min     0.497    0.565    0.257    0.120    0.218    0.095 
  (mpi-min)/min     0.497    0.565    0.257    0.120    0.218    0.095 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:28:45 EDT.