Subject: Re: Thread benchmarks
To: Andrew Doran <firstname.lastname@example.org>
From: Adam Hamsik <email@example.com>
Date: 09/28/2007 21:25:11
-----BEGIN PGP SIGNED MESSAGE-----
On Sep 28, 2007, at 7:50 PM, Andrew Doran wrote:
> Back in March I posted some MySQL benchmarks after we switched to a
> threading model in -current *. I've spent a lot of time tuning the
> library so I thought I'd post a followup. The original benchmark
> that I used
> (supersmack) now performs much better on -current that it did a few
> ago, so I picked something else this time: MySQL sysbench.
> Most of the sysbench runs that I've seen to date have sysbench
> running on
> the same machine as the database. That's a good test but with the
> of small installations and out-of-band activity, production setups
> look like that. So I ran sysbench itself on a seperate dual core
> Here are the results, comparing NetBSD 3 with NetBSD-current:
> And NetBSD-current compared to other systems:
Wow your numbers are impressive :). I like your work.
> Note this is stock NetBSD-current with FreeBSD's malloc()
> (jemalloc) in
> libc. I'll be merging that some time soon.
> With the vmlocking CVS branch and Mindaugas' new scheduler NetBSD
> around 500 TPS. There is a very gradual fall off in the number of TPS
> achieved as the number of connections begins to ramp up. I suspect
> could be due to a weakness somewhere in the network stack, so I'm
> that a bit of time spent profiling with large numbers of
> connections could
> yield good results.
> * http://mail-index.netbsd.org/tech-kern/2007/03/02/0005.html
Proud NetBSD user.
We program to have fun.
Even when we program for money, we want to have fun as well.
~ Yukihiro Matsumoto
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
-----END PGP SIGNATURE-----