PSTSWM Origin2000 Algorithm Comparison

Performance Studies using

PSTSWM


SGI/Cray Research Origin2000 Algorithm Comparison

(transpose LT experiment II-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
(assuming transpose FFT)
Number of Timesteps: 12
Results:

Transpose LT (2) (mpi)
Algorithm Comparison
  T42L1     T21L2     T42L2     T85L2     T85L1     T85L4  
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal algorithm   logtrans  alltoallv  srtrans  alltoallv  swtrans  swtrans 
  (alltoallv-min)/min     0.361    0.000    0.018    0.000    0.003    0.144 
  (generic-min)/min     0.805    0.119    0.018    0.082    0.029    0.007 

Transpose LT (2) (shmem)
Algorithm Comparison
  T42L1     T21L2     T42L2     T85L2     T85L1     T85L4  
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal algorithm   swtrans  swtrans  srtrans  swtrans  swtrans  srtrans 
  (generic-min)/min     0.458    0.283    0.092    0.151    0.095    0.079 

Transpose LT (2) (combined)
Communication Library Comparisons
  T42L1     T21L2     T42L2     T85L2     T85L1     T85L4  
  P=32     P=16     P=8     P=32     P=16     P=8  
  optimal library   shmem  shmem  shmem  shmem  shmem  shmem 
  (alltoallv-min)/min     0.527    0.245    0.036    0.055    0.023    0.145 
  (mpi-min)/min     0.122    0.245    0.018    0.055    0.021    0.000 


Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:07:31 EDT.