PSTSWM SP3-375 Communication Protocol Performance

Performance Studies using

PSTSWM

IBM SP3-375 Winterhawk II Protocol Performance

(distributed LT experiment A / ring sum algorithm)

Date/Person: December 12, 1999 / P. Worley
Platform: IBM SP3 at Oak Ridge National Laboratory (bobcat.ccs.ornl.gov):
     62 4-way Winterhawk II SMP nodes (375 MHz POWER3 with 8MB L2 cache)
Environment: AIX 4.3.3;   PSSP 3.1.1
Code Version: 6.6.4
Compilation Options: mpxlf -O3 -qarch=auto -qtune=auto -qcache=auto
Link Options: -bmaxdata:0x70000000
Environment Variables: export MP_SHARED_MEMORY="yes"
Math Library: ESSL
Communication Library: MPI over the switch using user space for internode communication
MPI using shared memory for intranode communication
Parallel Algorithm: ringsum
Number of Timesteps: 12
Results:

1x16 Processors / Problem T21L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  2.2643e-02   0.56   0.48   1.62 
Three Fastest
Protocols
1st2nd3rd
  a0   a6   c2 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   6   10 

1x32 Processors / Problem T42L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  4.2992e-02   0.64   0.56   1.89 
Three Fastest
Protocols
1st2nd3rd
  a0   a6   c4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   3   10 

1x8 Processors / Problem T42L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  6.0466e-02   0.31   0.29   0.73 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   d4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   9 

1x16 Processors / Problem T85L1
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.0884e-01   0.41   0.41   0.79 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   d4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  1   2   5 

1x32 Processors / Problem T85L2
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  1.9038e-01   0.47   0.48   0.94 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   d4 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   2   5 

1x8 Processors / Problem T85L4
Runtime Statistics
min(mean-min)/min(median-min)/min(max-min)/min
  5.4161e-01   0.07   0.03   0.84 
Three Fastest
Protocols
1st2nd3rd
  c4   c5   c3 
       Number of Proctocols With
Runtimes Within X% of Min
1%5%25%
  2   15   24 

Patrick H. Worley / ( worleyph@ornl.gov)
Last Modified Monday, 15-Jul-2002 10:19:27 EDT.