Re: RAID card recommendation

From: Glyn Astill <glynastill(at)yahoo(dot)co(dot)uk>
To: Jochen Erwied <jochen(at)pgsql-performance(dot)erwied(dot)eu>, Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: RAID card recommendation
Date: 2009-11-25 11:09:32
Message-ID: 498887.87084.qm@web23604.mail.ird.yahoo.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

--- On Tue, 24/11/09, Scott Marlowe <scott(dot)marlowe(at)gmail(dot)com> wrote:

> Jochen Erwied
> <jochen(at)pgsql-performance(dot)erwied(dot)eu>
> wrote:
> >
> > Since I'm currently looking at upgrading my own
> database server, maybe some
> > of the experts can give a comment on one of the
> following controllers:
> >
> > - Promise Technology Supertrak ES4650 + additional
> BBU
> > - Adaptec RAID 5405 SGL/256 SATA/SAS + additional BBU
> > - Adaptec RAID 5405Z SGL/512 SATA/SAS
> >
> > My personal favourite currently is the 5405Z, since it
> does not require
> > regular battery replacements and because it has 512MB
> of cache.
>
> Have you searched the -performance archives for references
> to them?
> I'm not that familiar with Adaptec RAID controllers. 
> Not requiring a
> battery check / replacement is nice.
>

We've been running Adaptec 5805s for the past year and I've been pretty happy, I think they have the same dual core IOP348 as the Areca 1680s.

I've a bunch of 5805Zs on my desk ready to go in some new servers too (that means more perc6 cards to chuck on my smash pile) and I'm excited to see how they go; I feer the unknown a bit though, and I'm not sure the sight big capacitors is reassuruing me...

Only problem I've seen is one controller periodically report it's too hot, but I suspect that may be something to do with the server directly above it having fanless power supplies.

In response to

Browse pgsql-performance by date

  From Date Subject
Next Message Richard Neill 2009-11-25 12:10:42 Re: Query times change by orders of magnitude as DB ages
Previous Message Matthew Wakeling 2009-11-25 11:07:36 Re: [PERFORM] Strange performance degradation