Compaq ES45 Benchmark

COMPAQ ES45 Benchmarks

Machine Description

1.PCI cards, 2.Fans, 3.Diskette drive, 4.Removable media bays, 5.CD-ROM drive, 6.4-bay disk drive storage cages, 7.Control panel, 8.Fans, 9.CPUs, 10.Memory.	Model Number	COMPAQ Alpha Server ES45
	Number of Processors	4
	Processor Type	Alpha EV68 at 1,000 MHz
	Operating System	Compaq UNIX
	Total Memory	1 Gbyte (2x512Mbyte)
	FMS Memory	730 Mbytes
	Number of Controllers	2
	Disks per Controller	Three 36 Gbyte disk

Problem Description

Code Used	Example 11
Matrix Sparsity	Full
Matrix Symmetry	Nonsymmetric with pivoting
Data Type	16-byte Complex

Factoring Results

Number of Equations	10,000	15,000	20,000	50,000
Date Started	10/29/2001	10/29/2001	10/29/2001	10/30/2001
CPU Time (Days::Hr:Min:Sec)	27:20	1:31:31	3:33:07	2::07:14:47
Wall Time (Hr:Min:Sec)	7:09	23:49	55:46	14:12:29
I/O Wait Time (Hr:Min:Sec)	0:05	0:14	1:18	4:27
Megaflops per Processor	1,626	1,639	1,668	1,676
Overall Megaflops	6,220	6,298	6,377	6,517
Parellel Speedup	3.82 out of 4	3.84 out of 4	3.82 out of 4	3.89 out of 4

Solving Results

Number of Equations	10,000	15,000	20,000	50,000
Number of Vectors	1,000	1,000	1,000	1,000
Date Started	10/29/2001	10/29/2001	10/29/2001	10/30/2001
CPU Time (Hr:Min:Sec)	8:13	17:50	31:33	3:26:52
Wall Time (Hr:Min:Sec)	2:07	4:35	8:06	55:02
I/O Wait Time (Hr:Min:Sec)	0:03	0:05	0:06	4:27
Megaflops per Processor	1,621	1,639	1,690	1,611
Overall Megaflops	6,292	6,298	6,579	6,056
Parellel Speedup	3.88 out of 4	3.84 out of 4	3.89 out of 4	3.76 out of 4

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. When reduced operation algorithms are used, this is based on the number of floating point operations that would have been performed by the traditional algorithm.

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

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

Copyright © 2001 Multipath Corporation