HP3000-L Archives

February 1996, 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:
Steve Cole <[log in to unmask]>
Reply To:
Date:
Mon, 12 Feb 1996 00:02:49 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (29 lines)
In a message dated 96-02-07 20:22:03 EST, [log in to unmask]
(Jim Viera) writes:
 
>The bottom line, your best bet is to run Lasser/Rx or equivalent tool to see
>whay your physical I/O rates are in your current disk configuration. You
>can run Glance or equivalent if you don't have a data collector running on
>your system. Then size accordingly......
>
>We are getting over 30 physical I/O's per second on our disc arrays
>before they reach the 100% utilization. (C2254HA)
>
>
 
My experience with the C2252HA and C2254HA disk arrays yeilded
the exact same 30 I/Os per second that Jim observed.  On systems with
higher I/O rates we were forced to use the C2252HA arrays to prevent the
I/O bottle necks.  On system with lower I/O rates the C2258HA arrays
worked fine.
 
In addition to LaserRX, GlanceXL or other on-line performance monitoring
tools will provide a good insight to your systems performance.  Keep in
mind that high I/O rates can be the caused by insufficient memory that
can cause excessive page faults or extremely fragmented disk.  If your
I/O rates are too high for the arrays may there is a reason that needs to
be addressed.
 
Steve Cole
Outer Banks Solutions, Inc.

ATOM RSS1 RSS2