IBM BladeCenter JS22 STREAM Standard Submission

From: Peter Wong <wpeter@us.ibm.com>
Date: Mon Nov 05 2007 - 17:07:09 CST

STREAM Editor's Note: On 2008-04-08, IBM requested that the model name of this system be updated to "IBM BladeCenter JS22 Express (4.0 GHz POWER6 SLES 10 SP1)".
I updated the name of this system in the tables to meet this request, except for the parenthetical element -- there is just not enough room in the tables for this verbose annotation.
Maybe someday I will update the database....


John,

I would like to submit this STREAM standard result for IBM BladeCenter JS22
with 4 cores, 16 GB (4 x 4 GB) DDR2 667MHz, four 4MB L2 caches and "no" L3
caches. Large pages (16MB) are used to back the arrays.

The following pieces of software are installed on the machine:
- Novell SUSE Linux Enterprise Server (SLES) 10 SP1 for IBM POWER
- libhugetlbfs 1.0.1 included with SLES 10 SP1
- IBM XL Fortran Advanced Edition V11.1 for Linux
- IBM XL C/C++ Advanced Edition V9.0 for Linux

Computer Array size SMT Triad score
------------------------------------- ----------- --- -----------
IBM BladeCenter JS22 (4.0 GHz POWER6) 105,000,000 on 15700.8364

Please make this result publicly available on or after November 6, 2007
in support of the announcement of IBM BladeCenter JS22.

Let me know if you have any questions.

Thanks,
Peter
--------------------------
Peter Wai Yee Wong, Ph.D.
IBM Linux Performance Team
wpeter@us.ibm.com

==========================================================================
----------------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
 Array size = 105000000
 Offset = 0
 The total memory requirement is 2403 MB
 You are running each test 10 times
 --
 The *best* time for each test is used
 *EXCLUDING* the first and last iterations
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 Number of Threads = 8
 ----------------------------------------------------
 Your clock granularity/precision appears to be 1 microseconds
 ----------------------------------------------------
Function Rate (MB/s) Avg time Min time Max time
Copy: 13938.8924 .1207 .1205 .1212
Scale: 13923.8812 .1207 .1207 .1209
Add: 15669.3901 .1609 .1608 .1611
Triad: 15700.8364 .1606 .1605 .1610
 ----------------------------------------------------
 Solution Validates!
 ----------------------------------------------------

Received on Tue Nov 06 10:36:03 2007

This archive was generated by hypermail 2.1.8 : Tue Nov 06 2007 - 14:44:42 CST