HomeGo Up

SGI Origin 300 Benchmarks

Machine Description

SGI Origin 300
Model Number SGI Origin 300
Number of Processors 8
Processor Type

MIPS R14000 at 500 MHz
Operating System

SGI IRIX64 Release 6.5
Total Memory

8 Gbytes
FMS Memory 1,792 Mbytes
Number of Controllers 2
Disks per Controller Two 73 Gbyte disks on one controller
One 73 Gbyte disk on the other controller

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 11/07/2001 11/07/2001 11/07/2001 11/08/2001
CPU Time (Days::Hr:Min:Sec) 55:13 2:58:50 7:29:18 4::13:10:58
Wall Time (Hr:Min:Sec) 7:32 23:58 59:02 14:10:53
I/O Wait Time (Hr:Min:Sec) 0:11 0:28 0:45 4:47
Megaflops per Processor 855 838 791 848
Overall Megaflops 5,900 6,259 6,023 6,529
Parellel Speedup 7.33 out of 8 7.46 out of 8 7.61 out of 8 7.70 out of 8

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 11/07/2001 11/07/2001 11/07/2001 11/08/2001
CPU Time (Hr:Min:Sec) 20:06 43:11 1:15:40 7:03:40
Wall Time (Hr:Min:Sec) 2:37 5:39 9:49 1:01:12
I/O Wait Time (Hr:Min:Sec) 0:30 0:20 0:34 0:50
Megaflops per Processor 663 694 704 698
Overall Megaflops 5,088 5,315 5,429 5,446
Parellel Speedup 7.67 out of 8 7.65 out of 8 7.70 out of 8 7.79 out of 8

NOTES:

  1. CPU Time is the total amount of User and System time used by all processors.
  2. 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.
  3. I/O Wait Time is the total time spent waiting for I/O to complete that is not overlapped by asynchronous I/O.
  4. 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.
  5. Overall Megaflops is the total number of floating point operations performed, divided by the Wall Time.
  6. Parallel Speedup is overall increase in performance due to parallel processing.


HomeGo Up
Copyright © 2001 Multipath Corporation