[OpenIndiana-discuss] Fileserver performance with log and cache devices?

[Geoff Flarity]

I've used filebench on solaris 10 for testing different disk
combinations for a database servers a head of a major purchase.  I had
to write my own graphing scripts, but the information was very
informative.

My advice would be to max out your RAM (for ARC) before you bother
with a ZIL, or L2ARC.  Where a fast SSD for a ZIL really shines is
random synchronized writes. IE a database transaction. You'll notice
this when you run filebench and look at the results. I'm not sure how
NFS handles FSYNC and DSYNC to be honest.

Cheers,
GF



On Thu, Dec 1, 2011 at 9:59 AM, Gary Mills <gary_mills at fastmail.fm> wrote:
> I'm in the process of setting up a fileserver running oi_151a.  How
> can I determine the performance improvement from adding log and cache
> devices?  I'm using filebench with the fileserver personality.  Is
> that my best choice?  So far, I've tried filebench directly on the
> server and also on an NFS client.  What statistics should I be
> watching?
>
> To begin, I'm using a single 1 TB SAS disk on the server, without log
> or cache devices.  I intend to add an Intel 311 SSD for the log and
> Intel 320 SSDs for the cache.  I'd like to ensure that these improve
> NFS performance.  I'd like CIFS and zfs send performance to be
> reasonable as well.
>
> --
> -Gary Mills-            -refurb-                -Winnipeg, Manitoba, Canada-
>
> _______________________________________________
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss