Home Go Up

SGI Altix

Machine Description

Model Number SGI Altix 350
Processor Type Itanium-2 @ 1.4GHz
Number of Processors 16
Memory Installed 64 Gbytes
Memory Used by FMS 12 Gbytes
Disk Configuration XFS File System
(8) 73 Gb drives
(4) Controllers
(8) FMS Stripes (NUMIOQ=8)
Direct I/O (FMS IOTYPE=2)
Operating System Linux 2.4.21-sgi240rp04051808_10074

Problem Description

Code Used Example 11
Matrix Sparsity Full
Matrix Symmetry Nonsymmetric
Data Type 16-byte Complex
Number of Equations 10,000 to 100,000
Number of Vectors 1
Reduced Operation Algorithms None (IALGOR=0)

Results for Complex Nonsymmetric Matrices

Date
Started		 Number of
Equations		 Number of
Processors		 Storage Times (Day::Hr:Min:Sec) Megaflops Parallel
Speedup
Loc. Gbytes CPU Wall I/O Wait CPU Overall
6/14/04 10,000 1 Disk 2.2 9:08 9:09 0:01 4,868 4,862 1.0
6/14/04 10,000 2 Disk 2.2 9:47 5:00 0:01 4,546 8,875 1.9
6/14/04 10,000 4 Disk 2.2 9:40 2:33 0:01 4,601 17,416 3.8
6/14/04 10,000 8 Disk 1.9 10:54 1:32 0:01 4,076 28,901 7.1
6/11/04 10,000 16 Disk 2.0 10:49 0:50 0:01 4,107 53,392 13.0
6/02/04 10,000 16 Mem. 2.0 11:14 0:53 0 3,959 50,123 12.7
6/14/04 20,000 1 Disk 7.9 1:21:12 1:21:13 0:04 4,379 4,378 1.0
6/14/04 20,000 2 Disk 8.6 1:14:46 39:18 0:04 4,756 9,047 1.9
6/14/04 20,000 4 Disk 8.6 1:15:34 19:28 0:04 4,705 18,259 3.6
6/14/04 20,000 8 Disk 8.6 1:19:03 10:43 0:08 4,498 33,169 7.4
6/11/04 20,000 16 Disk 7.5 1:24:37 6:00 0:14 4,202 59,221 14.1
6/02/04 20,000 16 Mem. 7.5 1:24:30 6:09 0 4,208 57,892 13.8
6/11/04 50,000 16 Disk 50.0 20:27:54 1:23:12 0:48 4,524 66,777 14.8
6/11/04 60,000 16 Disk 82.3 1::10:39:37 2:20:51 1:37 4,616 68,162 14.8
6/14/04 70,000 16 Disk 102.8 2::08:49:50 3:48:20 1:37 4,470 66,763 14.9
6/15/04 75,000 16 Disk 102.8 2::22:04:55 4:40:57 1:30 4,459 66,742 15.0
6/11/04 80,000 16 Disk 125.7 3::10:58:09 5:31:41 0:15 4,571 68,607 15.0
6/16/04 100,000 16 Disk 178.2 6::19:05:57 10:50:03 2:29 4,541 68,371 15.1

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.
  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 © Multipath Corporation