Home Go Up

IBM 650

Machine Description

Model Number 650
Processor Type Power4 @ 1.45GHz
Number of Processors 8
Memory Installed 16 Gbytes
Memory Used by FMS 12 Gbytes
Internal Disks (4) 36Gb; 3-way FMS striped
External Disks (12) 36Gb; Hardware striped
Operating System AIX 5.1
ESSL Version 03.03.00.06

Problem Description

Code Used Example 11
Matrix Sparsity Full
Matrix Symmetry Symmetric
Data Type 16-byte Complex
Number of Equations 20,000 to 200,000
Number of Vectors 1
Type of I/O Direct (IOTYPE=2)
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
7/30/03 20,000 Mem. 3.8 50:55 6:36 :00 3,491 26,911 7.71
7/30/03 20,000 Int.Disk 3.8 51:15 6:43 :05 3,469 26,501 7.64
7/31/03 40,000 Int.Disk 15.0 6:51:43 52:38 :15 3,454 27,019 7.82
8/01/03 60,000 Int.Disk 35.5 23:06:57 2:58:18 1:59 3,460 26,920 7.78
7/31/03 80,000 Int.Disk 60.9 2::06:36:01 6:57:35 3:23 3,473 27,247 7.85
7/30/03 100,000 Ext.Disk 93.0 4::09:35:31 13:24:08 1:49 3,507 27,635 7.88
8/01/03 200,000 Ext.Disk 355.0 34::20:17:52 4::09:15:51 5:56 3,543 28,148 7.94

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