Weekly Edition
Return to the Announcements pageRecent Features
| |||||||||||||
One Month Of Monitoring The Linux Kernel Performance (Phoronix)
The Phoronix folks have launched a project to track Linux kernel performance on a daily basis. The results for the first month are now available. "For those that may have forgot, at the start of December we launched the Phoronix Kernel Test Farm to begin benchmarking the Linux kernel on a daily basis using the automated tools that we provide via the Phoronix Test Suite and Phoromatic. Towards the middle of December we then unveiled the Phoromatic Tracker, which exposes these test results in real-time to the public. Well, it's now been a month of monitoring the kernel's performance and the entire Linux 2.6.33 kernel development cycle thus far, with many interesting findings."
(Log in to post comments)
One Month Of Monitoring The Linux Kernel Performance (Phoronix) Posted Jan 4, 2010 19:49 UTC (Mon) by jimparis (subscriber, #38647) [Link]
The "interesting findings" are basically "some stuff got better, some stuff got worse". Not very insightful.
One Month Of Monitoring The Linux Kernel Performance (Phoronix) Posted Jan 4, 2010 20:45 UTC (Mon) by arjan (subscriber, #36785) [Link]
indeed; they don't even identify which patch caused the change...
... this unlike the linux-kernel performance project which does something very similar but does bisect all the changes and even fixes issues ...
Yanmin and co do a really good job on that..
One Month Of Monitoring The Linux Kernel Performance (Phoronix) Posted Jan 5, 2010 5:01 UTC (Tue) by nikanth (guest, #50093) [Link]
Can be a good/bad advertisement for Linux, which could help indirectly. And if it gets popular, more tests would get added.
Should not be bad thing for Linux.
One Month Of Monitoring The Linux Kernel Performance (Phoronix) Posted Jan 5, 2010 9:21 UTC (Tue) by trasz (guest, #45786) [Link]
If only the Phoronix guys could get some basic knowledge about performance testing (average, standard deviation, information about the number of tests performed) and operating systems - most of the tests in PTS don't measure performance of the operating system at all, they either measure how well the compiler optimizes the code, or they are completelny phony, like the "threaded I/O tester", which doesn't actually measure I/O at all.
One Month Of Monitoring The Linux Kernel Performance (Phoronix) Posted Jan 5, 2010 10:08 UTC (Tue) by ErikA (guest, #53177) [Link]
Maybe you should inform Michael Larabel @ Phoronix about your feedback in
which tests that are used, instead of complaining here.
|
|||||||||||||