HP3000-L Archives

May 1999, 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:
Bill Lancaster <[log in to unmask]>
Reply To:
Bill Lancaster <[log in to unmask]>
Date:
Fri, 14 May 1999 00:51:28 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (33 lines)
Phil,

There isn't an absolute answer other than "It Depends" :-)

Generally, reducing spindles is a bad thing but it really depends on how
spindle-sensitive the application(s) is/are.  Also, there is only some
small amount of anecdotal information about AutoRAID on MPE so we can't
really know for sure.  The reduction of I/O parallelism *may* be offset by
the performance advantages of AutoRAID but I wouldn't bet the ranch on it,
especially since you are going to 6 spindles from 28.

If you are determined to be one of the first MPE AutoRAID users (a pretty
good bet, BTW, if you handle it right) I would recommend that you configure
the AutoRAID so as to lose as few spindles as possible as compared to your
current environment.  In order to do this, you may want to consider the
smallest drives available at the time you place the order.  Generally I
recommend that people stay at 4gb drives for OLTP data, though I don't know
what size of disks will be available with AutoRAID.

HTH,

Bill Lancaster

At 05:17 PM 5/12/99 -0500, Phil Anthony wrote:
>I guess I should have been more specific in my question:
>
>Is the effect on performance so adverse that I should consider limiting the
>reduction in spindles?  Or, as I have heard, is the loss of spindles offset
>by the speed and amount of buffering that goes on internally to the
>AutoRaid?
>
>

ATOM RSS1 RSS2