Re: RAID 5 vs RAID 10 benchmark

From: Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk>
Date: Sun, 28 Nov 2004 11:33:13 +0000 (UTC)
Message-ID: <cocd1p$kho$1@sparta.btinternet.com>

Notes in-line

Regards

Jonathan Lewis

http://www.jlcomp.demon.co.uk/faq/ind_faq.html The Co-operative Oracle Users' FAQ

http://www.jlcomp.demon.co.uk/seminar.html Optimising Oracle Seminar - schedule updated Sept 19th

"Paul Drake" <bdbafh_at_gmail.com> wrote in message news:910046b4.0411261612.5a6b1c3b_at_posting.google.com...
> see_at_reply-to.invalid (Bruno Jargot) wrote in message
> news:<1gnvkn8.59mfua1rzpu5sN%see_at_reply-to.invalid>...
>
> 16 GB of cache in front of 8 disks. Uh huh.
> I think that one could safely say that "the ROI would suck" -
> regardless of whether you have those disks arranged in a RAID 5 or
> RAID 10 config.
>

Remember that you can get something like 200GB on to a single disc, so that could be just one percent cache.

And, as one industry pundit is fond of saying at present, memory is very cheap, and the cheapest solution is often just to whack in a load more memory.

Personally, I find that there RAID-5 / RAID-10 argument is pretty pointless, the first issue is to persuade people that you can't put a 180GB database onto a single 200GB disc if you want any serious degree of activity on the database. Received on Sun Nov 28 2004 - 05:33:13 CST