PSTSWM Origin2000 Algorithm Comparison

Performance Studies using

PSTSWM

SGI/Cray Research Origin2000 Algorithm Comparison

(distributed LT experiment I-C )

Date/Person: May 11, 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:

Distributed LT (1) (mpi) / Column Major Processor Ordering
Algorithm Comparison
  T42L2     T42L2     T42L16    T85L4     T85L4     T85L32 
  P=2x32     P=4x16     P=8x8     P=2x32     P=4x16     P=8x8  
  optimal algorithm   halfsum  halfsum  halfsum  ringpipe  ringpipe  ringpipe 
  (allreduce-min)/min     0.478    0.167    0.128    2.203    0.457    0.439 
  (generic-min)/min     1.371    0.587    0.047    0.199    0.157    0.253 

Distributed LT (1) (shmem) / Column Major Processor Ordering
Algorithm Comparison
  T42L2     T42L2     T42L16    T85L4     T85L4     T85L32 
  P=2x32     P=4x16     P=8x8     P=2x32     P=4x16     P=8x8  
  optimal algorithm   halfsum  halfsum  ringpipe  ringpipe  ringpipe  ringpipe 
  (generic-min)/min     0.637    0.356    0.143    0.631    0.344    0.229 

Distributed LT (1) (combined) / Column Major Processor Ordering
Communication Library Comparisons
  T42L2     T42L2     T42L16    T85L4     T85L4     T85L32 
  P=2x32     P=4x16     P=8x8     P=2x32     P=4x16     P=8x8  
  optimal library   shmem  shmem  mpi  mpi  mpi  mpi 
  (allreduce-min)/min     0.660    0.561    0.128    2.203    0.457    0.439 
  (mpi-min)/min     0.123    0.338    0.000    0.000    0.000    0.000 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:06:11 EDT.