Messages in this thread | | | Date | Thu, 11 Feb 2010 16:49:35 -0500 | From | Rik van Riel <> | Subject | Re: [PATCH -mm] change anon_vma linking to fix multi-process server scalability issue |
| |
On 02/11/2010 04:04 PM, Andrew Morton wrote: > On Thu, 28 Jan 2010 00:20:00 -0500 > Rik van Riel<riel@redhat.com> wrote: > >> The old anon_vma code can lead to scalability issues with heavily >> forking workloads. Specifically, each anon_vma will be shared >> between the parent process and all its child processes.
Some test results:
Without the anon_vma changes, when AIM7 hits around 9.7k users (on a test box with 16GB RAM and not quite enough IO), the system ends up running >99% in system time, with every CPU on the same anon_vma lock in the pageout code.
With these changes, AIM7 hits the cross-over point around 29.7k users. This happens with ~99% IO wait time, there never seems to be any spike in system time. The anon_vma lock contention appears to be resolved.
-- All rights reversed.
| |