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:
Fri, 10 Dec 1999 12:57:56 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (23 lines)
> Good points. However, I have heard disk array sales people
> try to claim
> controller caching improves both read and write performance.

On systems that don't do effective read-ahead, it probably will help reads.
On MPE, read caching on the controller is probably wasted effort. I wonder
if MPE-oriented controllers can be configured to "write-only" caching? If
they used their cache only for writing (in "write-back" mode with power
protection), and bypassed the cache completely on reads, they might do even
better.

> In answer to
> your question in your other posting, the AutoRAID cache is 96
> MB, pretty
> small by today's standards.

Definitely small for higher-end hardware. Heck, even "PC class" hardware
like Mylex's PCI-card RAID controllers go up to 256MB!


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

ATOM RSS1 RSS2