[stream] Stream on SMP AMD Opteron

From: Jeffrey W. Baker <jwbaker@acm.org>
Date: Wed Sep 17 2003 - 18:54:20 CDT

Thanks for the Stream bench. These results are obtained on a RackSaver
rs1100, which is a 2-CPU AMD Opteron 244 running Linux 2.4.23, with gcc
3.3. The machine has 8GB main memory, 4GB local to each CPU, so it is a
slight NUMA. Compiler flags were -m64 -O.

-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 100000000, Offset = 0
Total memory required = -1807.2 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Your clock granularity/precision appears to be 9999 microseconds.
Each test below will take on the order of 689999 microseconds.
   (= 69 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
-------------------------------------------------------------
WARNING -- The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
-------------------------------------------------------------
Function Rate (MB/s) RMS time Min time Max time
Copy: 1797.7521 0.8970 0.8900 0.9000
Scale: 1818.1875 0.8830 0.8800 0.8900
Add: 2162.1610 1.1150 1.1100 1.1200
Triad: 2162.1684 1.1100 1.1100 1.1100

Cheers,
Jeffrey Baker
Received on Wed Sep 17 18:54:20 2003

This archive was generated by hypermail 2.1.8 : Tue Sep 23 2003 - 09:07:59 CDT