PSTSWM Origin2000 Algorithm Comparison

Performance Studies using

PSTSWM

SGI/Cray Research Origin2000 Algorithm Comparison

(transpose FFT experiment A)

Date/Person: May 8, 1998 / P. Worley
Platform: SGI Origin2000 at Los Alamos National Laboratory (nirvana11):
   128 195-MHz MIPS R10000 processors
Environment: IRIX 6.5
mpt_1.2.0.1
MIPSpro Compilers: Version MIPS pro_7.2
Code Version: 6.3
Compilation Options: f77 -64 -i4 -O3
Math Library: SCSL
Communication Library: MPI
SHMEM
Number of Timesteps: 12
Results:

Transpose FFT (mpi)
Algorithm Comparison
  T10L16    T10L16    T21L8     T21L32    T21L16    T42L16 
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal algorithm   logtrans  swtrans  srtrans  swtrans  swtrans  swtrans 
  (alltoallv-min)/min     0.073    0.022    0.085    0.084    0.099    0.281 
  (generic-min)/min     0.158    0.020    0.014    0.018    0.034    0.001 

Transpose FFT (shmem)
Algorithm Comparison
  T10L16    T10L16    T21L8     T21L32    T21L16    T42L16 
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal algorithm   swtrans  swtrans  srtrans  swtrans  swtrans  srtrans 
  (generic-min)/min     0.328    0.184    0.116    0.170    0.137    0.113 

Transpose FFT (combined)
Communication Library Comparisons
  T10L16    T10L16    T21L8     T21L32    T21L16    T42L16 
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal library   shmem  shmem  shmem  shmem  shmem  mpi 
  (alltoallv-min)/min     1.076    0.392    0.098    0.461    0.160    0.281 
  (mpi-min)/min     0.934    0.363    0.012    0.348    0.056    0.000 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:07:27 EDT.