From: Luke Lonergan <llonergan@greenplum.com>
To: Luke Lonergan <llonergan@greenplum.com>
To: Bruce Momjian <pgman@candle.pha.pa.us>
To: Christopher Browne <cbbrowne@acm.org>
Cc: pgsql-performance@postgresql.org
Subject: Re: Reliability recommendations
Date: Fri, 24 Feb 2006 07:41:12 -0800
Message-ID: <C0246918.1DB99%llonergan@greenplum.com> (raw)
In-Reply-To: <C02462CC.1DB8A%llonergan@greenplum.com>
Bruce,
On 2/24/06 7:14 AM, "Luke Lonergan" <llonergan@greenplum.com> wrote:
> So, if you want RAID5, these machines work for me. The lack of RAID 10
> could knock them out of contention for people.
Sorry in advance for the double post, but there's some more information on
this, which altogether demonstrates why people get frustrated with Dell IMO.
Later on in the article I cited, the same person who listed Dell's technical
description of their strange, not really RAID10 support, he mentions that he
called LSI directly to find out the real scoop. He says that they (LSI)
told him that the LSI controller does support real RAID10, so he concluded
that the Dell could do it too.
However, Dell's documentation seems unambiguous to me, and matches our
direct experience. Also, more online documentation from Dell reinforces
this. See their definition of both RAID10 and spanning in this
backgrounder:
http://support.dell.com/support/edocs/storage/RAID/RAIDbk0.pdf. Seems that
they are clear about what striping and spanning do, and what RAID10 does,
making their delineation of the standard RAID10 support versus what some of
their controllers do (all of the PERC 4 series) pretty clear.
So, I'd conclude at this point that Dell seems to have implemented a RAID
BIOS that does not allow true RAID10 or RAID50 on their embedded LSI
adapters, which could otherwise support RAID10/50. They have done this
intentionally, for some reason unknown to me and it seems to some people at
LSI as well.
- Luke
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: llonergan@greenplum.com, pgman@candle.pha.pa.us, cbbrowne@acm.org
Subject: Re: Reliability recommendations
In-Reply-To: <C0246918.1DB99%llonergan@greenplum.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox