Testing The Different Ubuntu 10.04 Kernels

Written by Michael Larabel in Software on 14 March 2010. Page 5 of 5. 19 Comments

The Dbench test when running 12 parallel clients produced a similar story to the IOzone write benchmark: the -preempt kernel did better than -server which did better than -generic, but the kernel did better than the Ubuntu kernels and the situation gets even better with the newer 2.6.33 kernel and it looks like it will be even better with the Linux 2.6.34 kernel. The deltas with Dbench were not as large with there being just an 18% advantage in favor of -preempt over -generic.

Next up was the FS-Mark disk test, but here it is a different outcome than Dbench and IOzone. The -generic kernel still ran slower than the -server and -preempt alternatives, but of the six kernels the mainline build did the best. There is a loss with the 2.6.33 kernel when running this 1000 files of 1MB size test, which is only partially recovered by the Linux 2.6.34-rc1 release. Using the -server kernel led to 8% more files per second being handled than with the default -generic kernel while the biggest gain was certainly with -preempt with 76% more files/sec over a stock Ubuntu installation.

The Bullet Physics Engine test did not lead to any significant performance differences in the kernels tested.

Lastly, with the simple dcraw test that converts high-resolution RAW image files to PPM, it was not really impacted by the kernels.

For the benchmarks that were influenced by the different kernels we tested, our simple summary is that the -preempt kernel had its advantages in the disk tests at least with our SSD-driven workstation, there are measurable ATI Radeon performance regressions in some OpenGL tests with the Ubuntu 2.6.32 kernels that are carrying the back-ported 2.6.33 DRM, the Ubuntu kernels are running well for the Apache server performance while our AMD Opteron Quad-Core setup experienced devastating results with this test on the newer post-2.6.32 kernels, and the PostgreSQL performance should be even better when it comes time for Ubuntu 10.10 or when switching to a newer kernel at least with the EXT4 file-system on a SSD. These tests can be carried out on your own system(s) in an easy, automated, and reproducible manner using the Phoronix Test Suite. Of course, leading up to the Ubuntu 10.04 LTS release next month we will be featuring this Linux operating system in more articles with benchmarks.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.

Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.