Progress on scaling of FreeBSD on 8 CPU systems

[Kris Kennaway]

On Sun, Feb 25, 2007 at 08:22:39PM +0100, Jeremie Le Hen wrote:
> Hi, Kris,
> 
> On Sat, Feb 24, 2007 at 04:55:08PM -0500, Kris Kennaway wrote:
> > Now that the goals of the SMPng project are complete, for the past
> > year or more several of us have been working hard on profiling FreeBSD
> > in various multiprocessor workloads, and looking for performance
> > bottlenecks to be optimized.
> > 
> > We have recently made significant progress on optimizing for MySQL
> > running on an 8-core amd64 system. The graph of results may be found
> > here:
> > 
> >   http://www.freebsd.org/~kris/scaling/scaling.png
> > 
> > This shows the graph of MySQL transactions/second performed by a
> > multi-threaded client workload against a local MySQL database with
> > varying numbers of client threads, with identically configured FreeBSD
> > and Linux systems on the same machine.
> 
> I'm really glad to be looking at these results eventually.  Thanks to
> all FreeBSD committers.
> 
> 
> > The test was run on FreeBSD 7.0, with the latest version of the ULE
> > 2.0 scheduler, the libthr threading library, and an uncommitted patch
> > from Jeff Roberson [1] that addresses poor scalability of file
> > descriptor locking (using a new sleepable mutex primitive); this patch
> > is responsible for almost all of the performance and scaling
> > improvements measured.  It also includes some other patches (collected
> > in my kris-contention p4 branch) that have been shown to help
> > contention in MySQL workloads in the past (including a UNIX domain
> > socket locking pushdown patch from Robert Watson), but these were
> > shown to only give small individual contributions, with a cumulative
> > effect on the order of 5-10%.
> 
> MySQL uses gettimeofday(2) very often.  ISTR it has been stated
> that MySQL performances could make the most of a system-wide shared
> page where the current time would be updated regularly by the kernel;
> gettimeofday(2) could consequentely work without any context switch.
> 
> Do the patches you're talking about include such a feature already ?

No, but when we tried hacking up gettimeofday last year to see what
effect the reduction of syscalls had, it didn't make a lot of
difference (at the time there was a theory that this was a major
source of the performance difference between freebsd and linux here).

It would still be worthwhile to explore implementing such a thing,
because it could give some incremental improvements, which are now
becoming more interesting to explore for this workload.

Kris
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/freebsd-performance/attachments/20070225/a90bc616/attachment.pgp