PSTSWM T3E-900 Point-to-Point Communication Performance

Performance Studies using

PSTSWM

SGI/Cray Research T3E-900 SWAP Performance

(unordered swap of 2MB message using SHMEM)

(performance measured per processor when all processors in node communicating)

Date/Person:	September 30, 1999 / 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 2.0.3.41
	mpt 1.2.1.3
	Cray CF90 Version 3.1.0.3
Communication Library:	SHMEM
SWAP size:	262144 REAL*8 floating point values each direction
Message size:	Largest - 262144 REAL*8 floating point values
	Smallest - 256 REAL*8 floating point values
Processors:	0 and 2
	1 and 3
Latency Definition:	(T₁₀₂₄-T₅₁₂)/512
Results:

unordered swap using nonblocking send

Data Statistics
bidirectional bandwidth estimated latency model error

(peak MByte/sec) (usec/msg) (max. rel. error)

1 iter. 554.40 8.49 8.7%

10 iter. 566.72 8.64 10.9%

1 iter. w/overlap 547.77 7.32 5.9%

10 iter. w/overlap 565.15 6.67 6.9%

unordered swap using nonblocking receive

Data Statistics
bidirectional bandwidth estimated latency model error

(peak MByte/sec) (usec/msg) (max. rel. error)

1 iter. 422.31 10.08 12.1%

10 iter. 427.50 10.10 12.8%

1 iter. w/overlap 419.25 8.33 7.7%

10 iter. w/overlap 426.75 7.78 7.8%

DISCUSSION

In these tests, we examine SHMEM communication performance between two pairs of processors, looking for signs of performance degradation due to network contention. A real test for network contention would use realistic communication patterns and more than four processors. However, this test is used as a control study for the tests on the systems using shared memory nodes, in which contention and other degradation occurs almost immediately.