henning@perfom.zko.dec.com (John Henning) on 06/16/2000 04:52:13 PM
To: John D Mccalpin/Austin/IBM@IBMUS
Subject: 1cpu Compaq_AlphaServer_GS320 and Compaq_AlphaServer_GS160
Hi John,
This is a single CPU on an AlphaServer GS160 or GS320.
This uses the same source code as
http://www.cs.virginia.edu/stream/stream_mail/2000/0013.html
but it was compiled without "-omp", so is single threaded.
This run was from the same day as the 32-CPU run.
- John
wfdemo1> f90 -fast -O5 -unroll 8 -arch ev6 mcc_omp_nodebug.f -o no_omp
wfdemo1> ./no_omp
n, offset, ntimes
120000549,0,10
----------------------------------------------
Double precision appears to have 16 digits of accuracy
Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
Array size = 120000549
Offset = 0
The total memory requirement is 2746.59 MB
You are running each test 10 times
The *best* time for each test is used
----------------------------------------------------
Your clock granularity/precision appears to be 1000 microseconds
The tests below will each take a time on the order
of 1430000 microseconds
(= 1430 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: 970.6819 2.6522 1.9780 3.5260
Scale: 958.5665 2.6117 2.0030 2.9400
Add: 1003.4889 3.7992 2.8700 4.9440
Triad: 1002.4411 3.7819 2.8730 4.4580
Sum of a is = 1.383967266434037E+020
Sum of b is = 2.767934540034517E+019
Sum of c is = 3.690579379856141E+019
This archive was generated by hypermail 2b29 : Sat Jun 17 2000 - 05:16:38 CDT