Home Go Up

SGI Altix

Machine Description

Model Number SGI Altix XE
Processor Type Intel X86-64 Quad Core @ 2.66GHz
Number of Processors 2
Number of Cores 8
Memory Installed 8 Gbytes
Memory Used by FMS 6 Gbytes
Operating System Linux 2.6.16.21-0.8-smp

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 100,000
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
04/06/07 20,000 Disk 4.0 25:18 3:32 0:07 7,027 50,357 7.2
04/06/07 40,000 Disk 17.0 3:18:12 26:56 1:08 7,176 52,798 7.4
04/06/07 60,000 Disk 31.2 11:07:36 1:26:42 0:57 7,190 55,367 7.7
04/06/07 80,000 Disk 56.7 1::01:40:01 3:19:26 2:14 7,388 57,053 7.7
04/06/07 100,000 Disk 88.6 2::01:24:12 6:22:00 3:51 7,497 58,175 7.8

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
04/06/07 20,000 Disk 8.0 49:26 6:40 0:14 7,193 53,281 7.4
04/06/07 40,000 Disk 31.0 6:35:56 52:07 1:27 7,184 54,588 7.6
04/06/07 60,000 Disk 66.5 21:34:24 2:47:14 2:01 7,416 57,405 7.7
04/06/07 80,000 Disk 115.2 2::02:12:50 6:26:20 3:25 7,553 58,902 7.8
04/06/07 100,000 Disk 177.2 4::01:37:48 12:29:14 4:48 7,587 59,321 7.8

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.
  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