Linux Benchmarking

Last updated: 12 October 2007

UnixBench has a new home!
UnixBench is now based at Google Code, and all maintenance will happen there. For the latest information and downloads, please go to: http://code.google.com/p/byte-unixbench/

This page presents some information on Linux benchmarking; and in particular, UnixBench Version 5, a version of the venerable Byte benchmarks updated to handle multi-CPU systems.

Results

Here are some benchmarking results from UnixBench. Click on a system name to see the full results:

System	CPU	OS	UnixBench Version	System		Graphics
System	CPU	OS	UnixBench Version	Single	Dual	2D	3D
HP Compaq nc8430	Core Duo T2500 @ 2.00GHz	openSUSE 10.2 (i586)	`5.0`	`678.2`	`1026.2`
HP Compaq nc8430	Core Duo T2500 @ 2.00GHz	openSUSE 10.3 (i586)	`5.0`	`495.7`	`802.7`
HP Compaq nc8430	Core Duo T2500 @ 2.00GHz	openSUSE 10.3 (i586)	`5.1`	`493.8`	`797.0`	`668.7`	`731.8`
Sony Vaio TXN25N	Core Solo U1500 @ 1.33GHz	openSUSE 10.2 (i586)	`5.0`	`498.4`	n/a
Sony Vaio TXN25N	Core Solo U1500 @ 1.33GHz	openSUSE 10.3 (i586)	`5.0`	`426.0`	n/a
Sony Vaio TXN25N	Core Solo U1500 @ 1.33GHz	openSUSE 10.3 (i586)	`5.1.1`	`437.5`	n/a	`730.9`	`209.6`
Sony Vaio TXN25N	Core Solo U1500 @ 1.33GHz	openSUSE 10.3 (i586) (runlevel 1)	`5.1.1`	`385.4`	n/a
Dell Latitude D620	Core 2 Duo T5600 @ 1.83GHz	openSUSE 10.3 (i586)	`5.0`	`560.1`	`1013.1`
Dell Latitude D620	Core 2 Duo T5600 @ 1.83GHz	openSUSE 10.3 (i586)	`5.1`	`575.2`	`1024.0`	`1080.2`	`267.1`
Dell Latitude D620	Core 2 Duo T5600 @ 1.83GHz	openSUSE 10.1 (X86-64)	`5.1.1`	`685.6`	`1186.1`	`1335.2`	`411.9`
Dell Latitude D620	Core 2 Duo T5600 @ 1.83GHz	openSUSE 10.2 (X86-64)	`5.1.1`	`684.8`	`1178.7`	`1243.2`	`405.6`
Dell Latitude D620	Core 2 Duo T5600 @ 1.83GHz	openSUSE 10.3 (X86-64)	`5.1.1`	`678.6`	`1125.6`	`1263.4`	`363.8`
HP Compaq 8510w	Core 2 Duo CPU T7500 @ 2.20GHz	openSUSE 10.3 (X86-64)	`5.1.1`	`842.5`	`1192.2`	`4886.9`	`3613.8`

Some observations:

OpenSUSE 10.3 is a bit slower than 10.2; it runs at about 75% - 85% as fast. Shame, since 10.2 was about 50% faster than 10.1.
Linux gets a good speedup from utilising 2 CPUs: 160% (Core Duo) to 180% (Core 2 Duo). Considering that the system benchmark covers a pretty broad range, including filesystem tests that are actually hurt by 2 CPUs (due to greater contention; see the "USAGE" file), this seems pretty good.
The 64-bit Linux install gains about 10%-15% over a 32-bit install on the same hardware.
UnixBench 5.1 discards the worst 1/3 of the results for each test, in an attempt to improve consistency. This causes a gain in the test scores of about 1%-2% relative to 5.0.
The graphics benchmarks, specially the 3D ones, are still very much in their infancy. Still, the Compaq — the only machine here with 3D hardware — clearly wins there.

Getting UnixBench

Please go to:

http://code.google.com/p/byte-unixbench/

Old UnixBench Versions

Version	Source	Documentation
5.1.2	unixbench-5.1.2.tar.gz	README	USAGE	WRITING_TESTS
5.1.1	unixbench-5.1.1.tar.gz	README	USAGE	WRITING_TESTS
5.1	unixbench-5.1.tar.gz	README	USAGE	WRITING_TESTS
5.0	unixbench-5.0.tar.gz	README	USAGE	WRITING_TESTS

Versions prior to 5.0 can be downloaded from David Niemi's site. Unixbench 5.0 was derived from version 4.1.0.

No Warranties: This information is provided free of charge and "as is" without any warranty, condition, or representation of any kind, either express or implied, including but not limited to, any warranty respecting non-infringement, and the implied warranties of conditions of merchantability and fitness for a particular purpose. All logos or trademarks on this site are the property of their respective owner. In no event shallthe author be liable for any direct, indirect, special, incidental, consequential or other damages howsoever caused whether arising in contract, tort, or otherwise, arising out of or in connection with the use or performance of the information contained on this web site.