Date: Mon, 3 Feb 2003 22:51:08 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: hackers@FreeBSD.ORG Cc: Tim Kientzle <kientzle@acm.org> Subject: More compartive power/performance results (was Re: Lower power SMP boxes?) Message-ID: <200302040651.h146p8Td041269@apollo.backplane.com> References: <200301312312.h0VNC5bQ007170@apollo.backplane.com> <3E3B1381.8050207@acm.org> <200302010150.h111oRFL007906@apollo.backplane.com> <3E3F438A.5040500@acm.org>
next in thread | previous in thread | raw e-mail | index | archive | help
test #1 test #2 Power
(seconds) (bandwidth) (load on UPS)
DELL2550 / 1.2 GHz 0.62 253MB/sec 12%
Pentium 4 / 1.3 GHz 0.74 500MB/sec ?
650 MHz Celeron P3 1.30 145MB/sec ?
EPIA M 9000 / 933 w/fan 1.55 72MB/sec 3.5-4%
EPIA 5000 / 533 fanless 2.50 65MB/sec 2%
test #1 (run several times so the file is cached)
/usr/bin/time -l sort /usr/share/dict/words > /dev/null
test #2 (run several times so the file is cached)
dd if=bigfile of=/dev/null bs=64k count=1024
Power: Difference in load percentage based on querying the UPS,
An APC Smart-UPS 1400 RM.
The Celeron is an HP Pavilion and the P4 is a Sony VAIO desktop. Sorry,
no power readings there, I can't shut down the sony and the HP is in
a different room. I've added EPIA M 9000 tests to the list (I just got
it in today). I'm actually surprised that the M 9000 doesn't have better
ram bandwidth with its DDR ram, but the cpu suds are a definite
improvement over the EPIA 5000 and power consumption is still reasonably
low. The EPIA 5000 still seems to have the best performance/power ratio
though.
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200302040651.h146p8Td041269>