*BSD News Article 12079


Return to BSD News archive

Path: sserve!manuel.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!haven.umd.edu!umd5!roissy.umd.edu!mark
From: mark@roissy.umd.edu (Mark Sienkiewicz)
Newsgroups: comp.os.386bsd.development
Subject: Re: Caching controllers
Message-ID: <18731@umd5.umd.edu>
Date: 1 Mar 93 16:57:57 GMT
References: <1993Feb25.092047.23360@netcom.com>
Sender: news@umd5.umd.edu
Organization: University of Maryland
Lines: 17

In article <1993Feb25.092047.23360@netcom.com> jmonroy@netcom.com (Jesus Monroy Jr) writes:
>        It may not be clear from that, but most hardware is designed
>   with MS-DOS's "fat file" system in mind and as such may in some
>   cases get confused...  There are other "cache controllers" that
>   just do a simple (or not so simple) read-ahead-strategy.  These
>   controllers are only duplicating the work of the
>   OS (Operating System).  Hence, AISI, no realy gain can be made
>   with "caching controllers".   In fact in most cases, like the
>   MS-DOS controllers, 386BSD is slowed to a silly pace.

Are you claiming that the caching disk controller is actually _slower_ to
respond than a non-caching controller?

But, unless the cache is really poorly designed, caching the wrong blocks 
shouldn't make the disk noticeably slower.  Do you have any hard data
on this?  (e.g. response times of various controllers with the same disks?)