Re: [squid-users] Benchmarks: ext3 vs. ReiserFS (fwd)

Hi Joe,

My squid is currently running ext2, which is very stable for 100-120 req/s.
To achieve better performance, am thinking of changing to ext3. But, from the
forum, there seems to be some performance issue on ext3. Am thinking of
testing both fs on kernel 2.4.19. Hope you can share what you have done...

- When was the last time you tested the performance of ext3 & reiserfs?
- How did you test the performance?
- What kind of workload you use?
- What kind of performance did your box achieve (req/s & response time)?

Thanks,
Wei Keong

On Tue, 15 Oct 2002, Wei Keong wrote:

>
>
> ---------- Forwarded message ----------
> Date: Sat, 13 Jul 2002 01:34:02 -0500
> From: Joe Cooper <joe@swelltech.com>
> To: Steve Snyder <swsnyder@insightbb.com>
> Cc: squid-users@squid-cache.org
> Subject: Re: [squid-users] Benchmarks: ext3 vs. ReiserFS
>
> Reconsider your faith in unrelated benchmarks instead, Steve. ;-)
>
> I've tested ext3 in all modes every few months for the past year. It is
> still significantly slower than ReiserFS for Squid workloads. ext3 is
> faster than ext2, but it sure isn't going to be replacing ReiserFS
> anytime soon (the difference is quite large). Not to mention that it
> was reliably oopsable when running Squid on it until very recently.
>
> I'm keeping an eye on it. The squid-users list will be the second to
> know if ext3 surpasses ReiserFS in performance and stability for Squid
> workloads (the first to know will be my clients... ;-).
>
> Steve Snyder wrote:
> > Interesting benchmarks of ext3 vs. ReiserFS:
> >
> > http://www.gurulabs.com/ext3-reiserfs.html
> >
> > This makes me reconsider my use of ReiserFS for my Squid cache.
>
> --
> Joe Cooper <joe@swelltech.com>
> Web caching appliances and support.
> http://www.swelltech.com
>
Received on Mon Oct 21 2002 - 22:21:05 MDT