STREAM Result on 4x PPro200Mhz/512KCache

From: Pekka Hedqvist (
Date: Mon Oct 21 1996 - 08:43:24 CDT

This STREAM results are from a Compaq ProLiant 5000 with
4 Pentium Pro at 200Mhz w. 512KCache each. It has 128MB
internal memory and runs Solaris 2.5.1.

The results seems very low. Lover than many Pentium based
machines and way lower than the Intel_Alder_PPro results -
which (I think) uses the same chipset.

This kind of machine is top of the line SuperServer from
Compaq and similar machines from ALR, HP etc. uses the
the same memory system etc.

Any idea of whats going on? One idea is that Compaq configurated
the memory wrongly and the 4-way interleawing doesn't work properly.

16.four-roses.[~/work/stream]: gcc -O4 stream_d.c second_cpu.c -lm -o str
17.four-roses.[~/work/stream]: ./str
This system uses 8 bytes per DOUBLE PRECISION word.
Array size = 2000000, Offset = 0
Total memory required = 45.8 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 369999 microseconds.
   (= 37 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: 57.1429 0.5600 0.5600 0.5600
Scale: 54.2373 0.5960 0.5900 0.6000
Add: 62.3377 0.7700 0.7700 0.7700
Triad: 58.5366 0.8280 0.8200 0.8300

This archive was generated by hypermail 2b29 : Tue Apr 18 2000 - 05:23:06 CDT