bench_sgi_o3k

SGI Origin 3000

Machine Description


			Model Number	Origin 3000
			Number of Processors	32, 64, 128, 224
			Processor Type	R12000 @ 400 MHz
			Operating System	IRIX64, Rel. 6.5

Problem Description

Code Used	Example 11
Matrix Sparsity	Full
Data Type	16-byte Complex

Results

Date Started	07/30/2000	07/30/2000	07/30/2000	07/31/2000	05/23/2001	05/27/2001
Number of Equations	15,000	15,000	15,000	50,000	100,000	200,000
Matrix Symmetry	Symmetric	Nonsymmetric	Nonsymmetric With Pivoting	Symmetric	Nonsymmetric With Pivoting	Nonsymmetric With Pivoting
Number of Processors	32	32	32	64	128	224
Memory for FMS (Gbytes)	6	6	6	12	24	42
CPU Time (Day::Hr:Min:Sec)	1:49:13	3:32:14	3:47:48	2::01:14:19	48::17:09:48	452::17:33:02
Wall Time (Day::Hr:Min:Sec)	4:48	7:42	10:03	1:14:19	10:09:37	2::06:53:19
I/O Wait Time (Hr:Min:Sec)	Incore	Incore	Incore	:06	6:55	13:29
Megaflops per Processor	686	706	658	714	633	545
Overall Megaflops	15,611	19,471	14,930	37,379	72,907	107,964
Parellel Speedup	22.73 out of 32	27.54 out of 32	22.67 out of 32	52.30 out of 64	115.07 out of 128	197.96 out of 224

NOTES:

CPU Time is the total amount of User and System time used by all processors.

Wall Time is the elapsed time measured on a dedicated machine. This includes the time spent processing and any time spent waiting for I/O to complete.

I/O Wait Time is the total time spent waiting for I/O to complete that is not overlapped by asynchronous I/O.

Megaflops per Processor is the theoretical number of floating point operations performed, divided by the CPU time.

Overall Megaflops is the total number of floating point operations performed, divided by the Wall Time.

Parallel Speedup is the overall increase in performance due to parallel processing.

Copyright © 2001 Multipath Corporation