From owner-freebsd-questions  Wed Nov 12 19:35:06 1997
Return-Path: <owner-freebsd-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id TAA08690
          for questions-outgoing; Wed, 12 Nov 1997 19:35:06 -0800 (PST)
          (envelope-from owner-freebsd-questions)
Received: from iectech.com (netgate.iectech.com [198.136.226.10])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA08671
          for <FreeBSD-questions@FreeBSD.ORG>; Wed, 12 Nov 1997 19:34:59 -0800 (PST)
          (envelope-from CPELTIER@iectech.com)
Received: by netgate.iectech.com id <6207>; Wed, 12 Nov 1997 18:14:17 -0500
From: Chris Peltier <CPELTIER@iectech.com>
To: "'FreeBSD-questions@FreeBSD.org'" <FreeBSD-questions@FreeBSD.ORG>
Subject: CCD and Squid
Date: Wed, 12 Nov 1997 22:25:38 -0500
X-Mailer:  Microsoft Exchange Server Internet Mail Connector Version 4.0.837.3
Encoding: 34 TEXT
Message-Id: <97Nov12.181417est.6207@netgate.iectech.com>
Sender: owner-freebsd-questions@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk


Hi,

I was wondering what CCD disk
strategy is best with Squid. Is it better to
just concatenate the drives or to stripe
across all drives. With Squids directory
cache structure can you take advantage 
of reducing overall disk access time by
using multiple spindles concatenated?
I would think that disk transfer rate is
not as important in this application and 
reducing access time by keeping logical
I/Os on a single spindle (instead of involving
them all) is. 

Does anybody have any thoughts or experience
with this?

I am building a Squid proxy with a 266MHZ PII,
512mb memory, and 3 - 9mb  Ultra wide, Seagate
Barracudas plus a system/swap disk. I am setting
this up as a parent server for our much smaller
siblings. 
FreeBSD 2.2.2 is the OS.


Sincerely, 
                        Chris Peltier 

*	email: CPELTIER@IECTECH.COM
*	voice: 215-257-4917
*	FAX:   215-257-4916