SGI Altix 3000, 1 GHz Itanium2, 16-CPU STREAM result

From: John Baron (jbaron@sgi.com)
Date: Tue Jan 07 2003 - 13:18:12 CST


SGI Altix(TM) 3000
1 GHz Itanium2(R)
3 MB L3 cache, 256 kB L2 cache
190 GB main memory
SGI ProPack(TM) v.2.1

Intel(R) Fortran Itanium(R) Compiler for Itanium(R)-based applications
Version 7.0, Build 20021210
compilation flags: -O3 -i8 -openmp -extend_source

The "dplace" placement tool was used to ensure that there were two OpenMP
threads per frontside bus, which matches the system architecture of two
processors per frontside bus.

----------------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
 Array size = 68799488
 Offset = 255
 The total memory requirement is 1574 MB
 You are running each test 50 times
 --
 The *best* time for each test is used
 *EXCLUDING* the first and last iterations
 ----------------------------------------------------
 Your clock granularity/precision appears to be 1 microseconds
 ----------------------------------------------------
Function Rate (MB/s) Avg time Min time Max time
Copy: 26789.0668 0.0411 0.0411 0.0412
Scale: 26514.2302 0.0415 0.0415 0.0416
Add: 31132.1633 0.0531 0.0530 0.0531
Triad: 31295.5644 0.0528 0.0528 0.0529
 ----------------------------------------------------
 Solution Validates!
 ----------------------------------------------------

-- 
John Baron                                              jbaron@sgi.com
SGI Performance Engineering and Math Libraries            650-933-7544


This archive was generated by hypermail 2.1.4 : Wed Jan 08 2003 - 09:19:55 CST