lkml.org 
[lkml]   [2008]   [Oct]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 0/4] pending scheduler updates

* Mike Galbraith <efault@gmx.de> wrote:

> > > With that patch, pgsql+oltp scales perfectly.
> >
> > hm, tempting.
>
> I disagree. Postgres's scaling problem is trivially corrected by
> twiddling knobs (or whatnot). [...]

okay, then we need to document it a bit more: what knobs need twiddling
to make it scale perfectly?

> [...] With that patch, you can't twiddle mysql throughput back, or
> disk intensive loads for that matter. You can tweak the preempt
> number, but it has nothing to do with lag, so anybody can preempt
> anybody else as you turn the knob toward zero. Chaos.

okay, convinced.

> > Have you tried to hack/fix pgsql to do proper wakeups?
>
> No, I tried to build without spinlocks to verify, but build croaked.
> Never went back to slogging through the code.

if it falls back to IPC semaphores that's a bad trade from a performance
POV. The best would be if it used proper futexes (i.e. pthread_mutex()
and friends) not some home-grown user-space spinlock thing.

Ingo


\
 
 \ /
  Last update: 2008-10-22 14:45    [W:0.038 / U:2.244 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site