Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 28 May 2001 18:08:53 +0000 (GMT)
From:      "E.B. Dreger" <eddy+public+spam@noc.everquick.net>
To:        hackers@FreeBSD.ORG
Subject:   Re: brainstorm: "intermediate" disk caching
Message-ID:  <Pine.LNX.4.20.0105281800050.6337-100000@www.everquick.net>
In-Reply-To: <3B128FB4.70AE7C69@i-clue.de>

next in thread | previous in thread | raw e-mail | index | archive | help
> Date: Mon, 28 May 2001 19:49:40 +0200
> From: Christoph Sold <so@server.i-clue.de>
> 
>> My gut feel is that this would be more trouble than it's worth, would
>> not net any overall performance*reliability (expressed as a
>> product) gain, and that one might actually realize a p*r decrease.
> 
> IMHO it would speed up your DB significantly to have it a) run on a RAID
> 10 array and b) have it run on the raw disk. Two layers of lag reduced
> (well, for reads it is possibly only one layer).

RAID 1+0 is what I use... but I was thinking of scalibility.  In a five-
drive array (using one as a hot spare), RAID 1+0 has 67% the capacity of
RAID 5.  More expensive per megabyte, but handles more db ops.

However, the numbers become less favorable with bigger RAID 1+0 arrays.

Also, intermediate caching *does not* inherently defeat the purpose of
RAID.  IO could be cached to a RAID 1 volume, then transferred to the RAID
5 volume... my question was if it was worth the hassle.

Of course, with 36 GB drives readily available, maybe I shouldn't worry
until I have a database larger than 72 GB. ;-)


Eddy

---------------------------------------------------------------------------

Brotsman & Dreger, Inc.
EverQuick Internet Division

Phone: (316) 794-8922

---------------------------------------------------------------------------

Date: Mon, 21 May 2001 11:23:58 +0000 (GMT)
From: A Trap <blacklist@brics.com>
To: blacklist@brics.com
Subject: Please ignore this portion of my mail signature.

These last few lines are a trap for address-harvesting spambots.  Do NOT
send mail to <blacklist@brics.com>, or you are likely to be blocked.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.20.0105281800050.6337-100000>