Horton

Re: Spinlocks, yet again: analysis and proposed patches

Mark Wong <markw@osdl.org>

View thread
From: Mark Wong <markw@osdl.org>
To: Douglas McNaught <doug@mcnaught.org>
Cc: Josh Berkus <josh@agliodbs.com>, pgsql-hackers@postgresql.org, Greg Stark <gsstark@mit.edu>, Tom Lane <tgl@sss.pgh.pa.us>, Marko Kreen <marko@l-t.ee>
Date: 2005-09-13T17:32:42Z
Lists: pgsql-hackers 
On Tue, 13 Sep 2005 12:21:45 -0400
Douglas McNaught <doug@mcnaught.org> wrote:

> Josh Berkus <josh@agliodbs.com> writes:
> 
> > Tom, All:
> >
> >> It seems to me what you've found is an outright bug in the linux scheduler.
> >> Perhaps posting it to linux-kernel would be worthwhile.
> >
> > For people using this on Linux 2.6, which scheduler are you using?   Deadline 
> > is the recommended one for databases, and does offer significant (+5-8%) 
> > benefits in some heavy-contention environments, at least in OSDL tests.
> 
> I thought 'deadline' was an I/O scheduler, not a CPU scheduler?

Yes, that's correct.  That's an i/o elevator algorithm.

Mark