*BSD News Article 21276


Return to BSD News archive

Newsgroups: comp.os.386bsd.questions
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!saimiri.primate.wisc.edu!news.doit.wisc.edu!psl.wisc.edu!128.104.200.15!ram
From: ram@xor.epi.wisc.edu (Ram Bhamidipaty)
Subject: Re: VLB IDE cache controller
In-Reply-To: yaot@hardy.u.washington.edu's message of 21 Sep 1993 08:46:40 GMT
Message-ID: <RAM.93Sep21145412@xor.epi.wisc.edu>
Sender: news@pslu1.psl.wisc.edu (USENET News System)
Organization: Physical Sciences Lab, UW-Madison
References: <27mf1g$ajt@news.u.washington.edu>
Date: 21 Sep 1993 19:54:12 GMT
Lines: 39

>>>>> ">" ==   <yaot@hardy.u.washington.edu> writes:
>> NNTP-Posting-Host: hardy.u.washington.edu

>> Hi there:

>> I just upgraded my old ISA IDE controller to
>> a VLB cache version.

>> ... However, after some thorough test, 
>> I notice that the cache controller doesn't seem to improve
>> the I/O performance at all.

>> Does Netbsd support VLB cache controller ? or I need some kind
>> of driver ? Or does cache controller help at all?


I'm no expert...but: 386bsd and its derivatives have a sophisticated
caching system that predict what blocks from the disk will be
needed and also try to optimize how the disk head moves across
the surface to minimize the amount of time spent waiting
for head movement (which dominates the amount of time disk accesses
take).  Note that DOS is a totally different case, since it does
not optimize head motion and does no caching on its own (I hear
that there are packages that do this, like smartdrive...)

Thus I would expect almost no benefit from a caching controller, since
the cache system in 386bsd is already doing that job.

A VLB controller might be able to speed transfer rates, but an IDE controller
can only handle two disks, which can be handled by the ISA bus. A SCSI
VLB controller on the other hand would be a big win when connecting several
drives.

-Ram
--
--------------
Ram Bhamidipaty                     use this ->ram@epidat.epi.wisc.edu
Department of Preventive Medicine   or this->ram%epidat.decnet@macc.wisc.edu
University of Wisconsin, Madison    Ignore the header. It might be wrong.