Path: utzoo!attcan!uunet!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!decwrl!elroy.jpl.nasa.gov!ucla-cs!uci-ics!truesdel
From: truesdel@ics.uci.edu (Scott Truesdell)
Newsgroups: comp.sys.mac
Subject: Re: Removable "HD"s -- are they worth it?
Message-ID: <20216@paris.ics.uci.edu>
Date: 29 Jul 89 04:08:51 GMT
References: <9095@venera.isi.edu> <3000@blake.acs.washington.edu>
Sender: news@paris.ics.uci.edu
Lines: 32

gwangung@blake.acs.washington.edu (Roger Tang) writes:

>	I'm using mine as a primary storage device currently.  This is working
>out fairly well for the time being, though I haven't had it for too long.
>THey're are definitely ideal as a second or backup HD, and my experience
>seems to say that they don't do too badly as a primary.

Reports are starting to surface on CompuServe and elsewhere about the 
inadvisability of using the SyQuests as a primary drive. It appears 
that they really aren't designed for the extended duty-cycle of a 
primary drive and the SyQuest mechanism itself can fail prematurely.


>	By the way, I've heard that the PLI turbo cache (which PLI sells
>separately) can lower access times down to 12 ms.  I'm not sure how much
>to believe the time, but I can believe that the PLI software can speed up
>the drive even more.

I am sceptical of most disk caching schemes when used as an advertising
ploy to pull down the "effective access times".  At best, a disk
caching scheme can be optimized for a few specific file sizes. I like
using SCSI Evaluator for comparing disk performance because it measures
averaged sustained throughput of various sized files. The "ringers"
really stand out on the graphic display. Under some caching schemes,
throughput actually decreases with cache enabled except for a sharp
peak at certain file sizes. I'm more interested in a nice flat curve of
predictable straight-line performance.

  --scott

--
Scott Truesdell