(no subject)

From: Bob Bomar <bob@bomar.us>
Date: Thu Aug 07 2003 - 19:50:56 CDT

I came across the STREAM Benchmark utility and
wanted to share my findings:

Machine 1:
Athlon XP 2200+
Gigabyte GA-7VAXP VIA KT400 Motherboard
512 MB ECC- Reg Corsair PC2100 266mhz RAM

-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 1000000, Offset = 0
Total memory required = 22.9 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Your clock granularity/precision appears to be 7812 microseconds.
Each test below will take on the order of 15625 microseconds.
   (= 2 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: 1024.0000 0.0214 0.0156 0.0234
Scale: 682.6667 0.0234 0.0234 0.0234
Add: 1024.0000 0.0291 0.0234 0.0312
Triad: 1024.0000 0.0299 0.0234 0.0312

OS:
[bob@warrior] ~>uname -a
FreeBSD warrior.bomar.us 4.8-STABLE FreeBSD 4.8-STABLE #2: Mon Aug 4 23:56:47 CDT 2003 root@warrior_4.bomar.us:/usr/obj/usr/src/sys/Warrior i386

Compiled the sourec from the ports collection.

Machine #2:
Alpha EV4 500mhz
Digital PC164 Motherboard
64 Mb ECC-Reg RAM

-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 1000000, Offset = 0
Total memory required = 22.9 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 99999 microseconds.
   (= 10 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: 114.2858 0.1452 0.1400 0.1600
Scale: 114.2858 0.1421 0.1400 0.1500
Add: 126.3158 0.1991 0.1900 0.2100
Triad: 126.3159 0.1980 0.1900 0.2000

OS:
[bob@commanche] ~>uname -a
FreeBSD commanche.bomar.us 4.8-STABLE FreeBSD 4.8-STABLE #0: Wed Jul 26 17:19:33 CDT 2023 root@commanche.bomar.us:/usr/obj/usr/src/sys/Commanche alpha

Machine #3:
PII 400Mhz
196Mb 100 Mhz RAM
Abit BH6 motherboard

-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 1000000, Offset = 0
Total memory required = 22.9 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Your clock granularity/precision appears to be 7812 microseconds.
Each test below will take on the order of 39062 microseconds.
   (= 5 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: 341.3333 0.0485 0.0469 0.0547
Scale: 409.6000 0.0511 0.0391 0.0547
Add: 384.0000 0.0688 0.0625 0.0703
Triad: 307.2000 0.0844 0.0781 0.0859

-- 
Bob Bomar
bob@bomar.us
http://www.bomar.us/
-----------------------------
FreeBSD: The Power to Serve
http://www.freebsd.org

Received on Thu Aug 7 19:50:56 2003

This archive was generated by hypermail 2.1.8 : Mon Aug 25 2003 - 10:20:32 CDT