stream results for compaq armada 1700

Dear John,

Sorry for sending to two addresses, I wasn't sure which one is up to date.

I'm not sure if you are interested as this laptop is very out-of-date, but in case you are below are the results from running stream on compaq armada 1700.

%dmesg
Copyright (c) 1992-2005 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
        The Regents of the University of California. All rights reserved.
FreeBSD 6.0-RELEASE #0: Thu Nov 3 09:36:13 UTC 2005
    root@x64.samsco.home:/usr/obj/usr/src/sys/GENERIC
Timecounter "i8254" frequency 1193182 Hz quality 0
CPU: Pentium II/Pentium II Xeon/Celeron (233.34-MHz 686-class CPU)
  Origin = "GenuineIntel" Id = 0x652 Stepping = 2
  Features=0x183f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PA
T,PSE36,MMX,FXSR>
real memory = 167739392 (159 MB)
avail memory = 154624000 (147 MB)
[skip]

FreeBSD has a stream port in its ports collection. I send you three sets of results which were obtained with stream compiled:
a) from the ports collection.
b) from the latest stream.f with default settings.
c) from the latest stream.f with n=4,000,000.
The results differ slightly.

a) stream compiled from the ports collection

%more /usr/ports/benchmarks/stream/Makefile
# New ports collection makefile for: stream
# Date created: 10 May 2001
# Whom: Scott Flatman <sf@dsinw.com>
#
# $FreeBSD: ports/benchmarks/stream/Makefile,v 1.8 2005/02/17 08:43:36 vs Exp $

PORTNAME= stream
PORTVERSION= 1.0
PORTREVISION= 2
CATEGORIES= benchmarks
MASTER_SITES= ftp://ftp.cs.virginia.edu/pub/stream/Code/
DISTFILES= stream.c README
DIST_SUBDIR= stream-${PORTVERSION}

MAINTAINER= sf@slappy.org
COMMENT= Synthetic benchmark program that measures sustainable memory ban
dwidth

PLIST_FILES= bin/stream
PORTDOCS= stream.c README

do-extract:
        @${MKDIR} ${BUILD_WRKSRC}
        @${CP} ${_DISTDIR}/stream.c ${BUILD_WRKSRC}

do-build:
        (cd ${BUILD_WRKSRC} ; ${CC} ${CFLAGS} -o stream stream.c -lm)

do-install:
        (cd ${BUILD_WRKSRC}; ${INSTALL_PROGRAM} stream ${PREFIX}/bin)
. if !defined(NOPORTDOCS)
                ${MKDIR} ${DOCSDIR}
                (cd ${_DISTDIR}; ${INSTALL_DATA} ${PORTDOCS} ${DOCSDIR})
. endif

.include <bsd.port.mk>
%

My gcc is:

%gcc --version
gcc (GCC) 3.4.4 [FreeBSD] 20050518
Copyright (C) 2004 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

%stream
-------------------------------------------------------------
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 6 microseconds.
Each test below will take on the order of 147367 microseconds.
   (= 24561 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) Avg time Min time Max time
Copy: 177.3453 0.1632 0.1804 0.1841
Scale: 177.0292 0.1630 0.1808 0.1815
Add: 204.9059 0.2113 0.2343 0.2357
Triad: 193.9041 0.2233 0.2475 0.2490
-------------------------------------------------------------
Solution Validates
-------------------------------------------------------------
%

b) stream compiled from source with default settings.

The following files from http://www.cs.virginia.edu/stream/FTP/Code/ were used:
mysecond.c 12-Sep-2003 20:07 1k
stream.f 03-Oct-2005 20:21 15k

compiled as suggested in your FAQ:

%gcc -c -DUNDERSCORE mysecond.c
%f77 -O stream.f mysecond.o -o stream

(By the way your compilation instruction has mysecond.c, not *.o in the second line.)

%./stream
----------------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
 ----------------------------------------------
 STREAM Version $Revision: 5.6 $
 ----------------------------------------------
 Array size = 2000000
 Offset = 0
 The total memory requirement is 45 MB
 You are running each test 10 times
 --
 The *best* time for each test is used
 *EXCLUDING* the first and last iterations
 ----------------------------------------------
 ----------------------------------------------
 Printing one line per active thread....
 ----------------------------------------------------
 Your clock granularity/precision appears to be 3 microseconds
 ----------------------------------------------------
Function Rate (MB/s) Avg time Min time Max time
Copy: 177.1391 0.1809 0.1806 0.1813
Scale: 176.7088 0.1816 0.1811 0.1825
Add: 207.1662 0.2323 0.2317 0.2331
Triad: 180.5482 0.2662 0.2659 0.2669
 ----------------------------------------------------
 Solution Validates!
 ----------------------------------------------------
%

c) compiled from latest source with n=4,000,000.

compiled as before

%./stream
----------------------------------------------
 Double precision appears to have 16 digits of accuracy
 Assuming 8 bytes per DOUBLE PRECISION word
----------------------------------------------
 ----------------------------------------------
 STREAM Version $Revision: 5.6 $
 ----------------------------------------------
 Array size = 4000000
 Offset = 0
 The total memory requirement is 91 MB
 You are running each test 10 times
 --
 The *best* time for each test is used
 *EXCLUDING* the first and last iterations
 ----------------------------------------------
 ----------------------------------------------
 Printing one line per active thread....
 ----------------------------------------------------
 Your clock granularity/precision appears to be 3 microseconds
 ----------------------------------------------------
Function Rate (MB/s) Avg time Min time Max time
Copy: 178.2312 0.3594 0.3591 0.3596
Scale: 177.6845 0.3610 0.3602 0.3618
Add: 208.6884 0.5129 0.4600 0.8383
Triad: 175.7296 0.5525 0.5463 0.5771
 ----------------------------------------------------
 Solution Validates!
 ----------------------------------------------------
%

Clock precision is estimated to be 6 microseconds in case (a) and as 3 microseconds in cases (b) and (c).

I found the results tables on your site very helpful.

yours
anton


--
Anton Shterenlikht
Room 2.6, Queen's Building
Mech. Eng. Dept.
Bristol University
University Walk, Bristol BS8 1TR, UK
Tel: +44 (0)117 928 8233
Fax: +44 (0)117 929 4423

Received on Thu Jun 29 12:57:32 2006