HP3000-L Archives

December 1999, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Steve Dirickson <[log in to unmask]>
Reply To:
Steve Dirickson <[log in to unmask]>
Date:
Thu, 9 Dec 1999 20:39:10 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (33 lines)
> Regardless of any theoretical RAID performance characteristics and any
> claims made by any manufacturer, I believe the best advice for someone
> considering RAID is to get a demo unit of whichever product
> and see how it performs on YOUR system with YOUR applications.

Definitely; numbers on a spec sheet don't bring in the dollars--performance
does.

> Disc arrays also promise to
> improve performance through the use of caching.  The caching
> in arrays is
> usually found to be of little or no value in improving overall system
> performance in the case of MPE, because MPE already does an
> excellent job of caching.

I think we're talking about different caching here. The big advantage of
RAID-controller caching is on writes, not reads. An intelligent RAID
controller (with power-fail protection for its cache) can accept a write
from the host and immediately acknowledge it as completed, even though
nothing has actually been sent to any disk. As you mention, intelligent OS
read-ahead caching will always give you more bang for the buck than
controller caching, because the pre-read is done "in context" to what is
actually going on in the application--something the controller simply can't
know about.

Safety note: *never* configure any disk cache setup--RAID or not--to "write
back" unless you have power-loss protection for it!



Steve Dirickson   WestWin Consulting
[log in to unmask]   (360) 598-6111

ATOM RSS1 RSS2