Intel Linux 3.3 To Linux 3.13 Kernel Benchmarks

[phoronix]

Phoronix: Intel Linux 3.3 To Linux 3.13 Kernel Benchmarks

The latest kernel benchmarking that happened at Phoronix was testing every major Linux kernel release from Linux 3.3 through the latest stable Linux 3.13 release from an Intel Sandy Bridge system to see how the kernel performance has evolved during the hardware's lifetime for key subsystems.

http://www.phoronix.com/vr.php?view=19813

[halo9en]

"Interestingly this system does not show any Linux 3.13 disk/block regression like we have seen on other systems with this new kernel".

And yet 3.13 sucks on that system too, when compared to 3.12.

[tillschaefer]

Test Setup

There is one question that arises for me whenever i see such outliers as for kernel 3.7. How does the test setup look like?

Are the tests repeated over different runs? If so, is there a variance? Do you user HyperBoost? If so, is there enough time for the laptop tp cool down between the tasks, ...

Maybe i missed this, but a general article about how the tests are done would be very helpful.

[Michael]

There is one question that arises for me whenever i see such outliers as for kernel 3.7. How does the test setup look like?

Are the tests repeated over different runs? If so, is there a variance? Do you user HyperBoost? If so, is there enough time for the laptop tp cool down between the tasks, ...

Maybe i missed this, but a general article about how the tests are done would be very helpful.

It's all handled via Phoronix Test Suite. If you look at the test profile at OpenBenchmarking.org or run PTS yourself it will show you how many runs are by default, PTS auto-increases run count if there's variance, etc etc.