HP3000-L Archives

May 2000, 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:
Joseph Rosenblatt <[log in to unmask]>
Reply To:
Joseph Rosenblatt <[log in to unmask]>
Date:
Tue, 9 May 2000 09:01:20 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (43 lines)
When I was running JBOD I was having problems do to power outages. Once the
discs stop spinning they would not start back up. As I recall these were
mostly the full height 3010 models. I think they were also defective in that
they leaked oil on to the discs and corrupted the disc. {They threw oil, hmm
so that's what happened to those British disc drives. ; ) } I would estimate
that our downtime during that year was 95% due to disc failure.

I have used EMC disc arrays in the past. After the initial install I did not
experience any down time due to disc failure. If you are still running SE
SCSI you may even see a performance gain when you go to FW SCSI.  Whether
you run RAID 5 (or RAID s as EMC calls it) or RAID 1, mirroring, will also
effect performance.

I currently run an HP array and have also not experienced any down time in
the last 18 months. As far as performance I don't have much user traffic so
it is hard to judge.

I hope this helps.
-----Original Message-----
From:   HP-3000 Systems Discussion [mailto:[log in to unmask]] On Behalf
Of Patrick Edwards
Sent:   Monday, May 08, 2000 11:29 PM
To:     [log in to unmask]
Subject:        Downtime % with JBOD disks

My organization is looking to upgrade our JBOD scsi disk arrays to some
sort of high-availability array.  For the presentation, I'm required to
present how our shop compares to the "industry standard" as far as downtime.

I'm looking for feedback from other shops regarding what % your shop is
down for:
1) disk drive crash  (we're at 85% for past 1 1/2)
2) OS or System software failures (12%)
3) Application or programming lockups (~3%)

If you are already on a high-availability solution, I'd really appreciate
any feedback about % performance hits (or speed) from EMC, Nike, XP256,
etc...

Thanks mucho!!
Patrick Edwards
Little Rock, AR

ATOM RSS1 RSS2