HP3000-L Archives

November 1997, 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:
"Gary L. Paveza, Jr." <[log in to unmask]>
Reply To:
Gary L. Paveza, Jr.
Date:
Mon, 10 Nov 1997 07:24:27 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (27 lines)
At 07:55 AM 11/8/97 GMT, MTX MISMgr wrote:
>>

<batter about using 4 gig drives snipped out>
>As far as reducing your 'spindle count' from 6 to 4, this can have a negative
> effect on performance depending on your environment, everything else being
> equal, more spindles is generally better.

Hmm..not sure I really agree with this totally.  Yes, by increasing your
spindle count on a volumeset you increase your i/o capability.  But, you
also increase your points of failure.  Now, on all our other volumesets, I
would be more interested in increasing spindle count, due to the fact that
we mirror those drives and thus, a failure isn't a problem.  But on the
MPEXL_SYSTEM_VOLUME_SET, I would MUCH rather have the spindle count low to
reduce the number of drives that can fail.  You lose one of those drives
and you end up reinstalling, and reloading all your data.  If HP would give
us the ability to mirror that set then I'd be more than happy to use more
smaller drives instead of a few large ones (HP - that would be another vote
for mirroring of the system drives).
**************************************************************
*** All opinions are mine - because no one else wants them ***
*** Gary L. Paveza, Jr.                                    ***
*** Technical Support Specialist                           ***
*** Decision Research Corporation                          ***
*** [log in to unmask]                                   ***
**************************************************************

ATOM RSS1 RSS2