Date: Mon, 28 May 2001 18:37:07 -0700 From: Mike Smith <msmith@freebsd.org> To: "E.B. Dreger" <eddy+public+spam@noc.everquick.net> Cc: hackers@FreeBSD.ORG Subject: Re: brainstorm: "intermediate" disk caching Message-ID: <200105290137.f4T1b7B05693@mass.dis.org> In-Reply-To: Your message of "Tue, 29 May 2001 01:23:28 -0000." <Pine.LNX.4.20.0105290122130.10997-100000@www.everquick.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> > Date: Mon, 28 May 2001 19:01:37 -0500 (CDT)
> > From: David Scheidt <dscheidt@tumbolia.com>
>
> [ snip ]
>
> > If you're really interested in database performance, remember "Spindles
> > is good." Spreading your IO load over as many seperate disks, on as
> > many independent IO channels as practical will improve performance.
>
> There's the problem. With RAID 1+0, total the storage of the spindles,
> and half of that is usable. With RAID 5, it asymptotically approaches
> unity.
... at the same time as your statistical reliability approaches zero.
Running more than a small number of disks in your RAID 5 stripe is unwise;
typically you'd use RAID 50 (stripes across a set of RAID 5 units, each of
maybe 5-10 disks each).
Having said all that, RAID 10 is a real performer just now courtesy of
3ware's product offerings; the capacity loss is offset by the cost
savings. I'm sure that the storage market will shift again.
--
... every activity meets with opposition, everyone who acts has his
rivals and unfortunately opponents also. But not because people want
to be opponents, rather because the tasks and relationships force
people to take different points of view. [Dr. Fritz Todt]
V I C T O R Y N O T V E N G E A N C E
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?200105290137.f4T1b7B05693>