Home Go Up

SGI Altix

Machine Description

Model Number SGI Altix 350
Processor Type Itanium-2 @ 1.6GHz
Number of Processors 16
Memory Installed 64 Gbytes
Memory Used by FMS 29 Gbytes
Operating System Linux 2.4.21-sgi304r1
Disk Configuration XFS File System
(8) 36 Gb drives
(4) Controllers
(8) FMS Stripes
(NUMIOQ=4); (NUMSTP=2)
Direct I/O (FMS IOTYPE=2)

Problem Description

Code Used Example 11
Matrix Sparsity Full
Matrix Symmetry Symmetric and Nonsymmetric
Data Type 16-byte Complex
Number of Equations 20,000 to 150,000
Number of Vectors 1
Reduced Operation Algorithms None (IALGOR=0)

Results for Complex Symmetric Matrices

Date
Started		 Number of
Equations		 Storage Times (Day::Hr:Min:Sec) Megaflops Parallel
Speedup
Loc. Gbytes CPU Wall I/O Wait CPU Overall
02/19/05 20,000 Mem. 3.6 33:17 2:30 0:00 5,340 71,104 13.3
02/19/05 20,000 Disk 3.6 33:25 2:31 0:01 5,319 70,577 13.3
02/19/05 40,000 Mem. 16.7 4:15:37 17:33 0:00 5,564 81,028 14.6
02/19/05 40,000 Disk 16.7 4:16:05 17:38 0:06 5,554 80,631 14.5
02/19/05 60,000 Disk 46.5 14:33:58 58:26 0:49 5,492 82,157 15.0
02/19/05 80,000 Disk 65.1 1::10:27:01 2:16:16 1:15 5,504 83,503 15.2
02/19/05 100,000 Disk 111.5 2::17:38:33 4:18:03 4:00 5,642 86,118 15.3
02/20/05 150,000 Disk 204.4 9::04:20:25 14:12:54 5:37 5,673 87,936 15.5

Results for Complex Nonsymmetric Matrices

Date
Started		 Number of
Equations		 Storage Times (Day::Hr:Min:Sec) Megaflops Parallel
Speedup
Loc. Gbytes CPU Wall I/O Wait CPU Overall
02/18/05 20,000 Mem. 7.1 1:04:20 4:19 0:00 5,528 82,368 14.9
02/18/05 20,000 Disk 7.1 1:04:46 4:26 0:05 5,491 80,163 14.6
02/18/05 40,000 Disk 33.5 8:23:02 33:16 0:27 5,654 85,504 15.1
02/18/05 60,000 Disk 92.9 1::04:38:09 1:54:00 3:39 5,587 84,216 15.1
02/19/05 80,000 Disk 130.1 2::20:00:15 4:27:41 5:23 5,577 85,010 15.2
02/19/05 100,000 Disk 223.0 5::09:26:56 8:24:10 10:43 5,722 88,157 15.4

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