ADSM-L

Re: [ADSM-L] Question concerning large filepools on HP EVA

2011-12-07 12:51:44
Subject: Re: [ADSM-L] Question concerning large filepools on HP EVA
From: Stefan Folkerts <stefan.folkerts AT GMAIL DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Wed, 7 Dec 2011 18:48:11 +0100
..."After talking to HP, the advice they got was to try to reduce the load
on the box since it shouldnt be used more than 30% of the time during 24
hours."...

This is just absurd but I have heard bad stories about the eva series
regarding maximum loads before, but that was more around the >70%
utilization being a problem if I recall correctly, all sort of strange
things can happen when you get above a certain load..70 or 80% I believe.
I have IBM DS3524/EXP3512 entry level machines running 3TB SAS disks
(filepool) and 300GB SAS disks (db/log) in ~90TB TSM configs and all is
just fine, busy all day but no issues and this is a 12k head and 10k
expansion draws so nothing to expensive here.


On Wed, Dec 7, 2011 at 4:35 PM, Prather, Wanda <wPrather AT icfi DOT com> wrote:

> Ditto what Howard said.  If you even told them what you were using it for,
> they should take it back.
> But in the short term, can you do the dedup in your front-end SAS pool?
> If stuff is deduped to start with, it will stay deduped when it migrates
> to the next pool.
> Should greatly reduce the I/O load on the target pool.
>
>
>
> -----Original Message-----
> From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf 
> Of
> Howard Coles
> Sent: Wednesday, December 07, 2011 9:59 AM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Re: [ADSM-L] Question concerning large filepools on HP EVA
>
> We run the XIV from IBM, which is similar, hardware wise, and we have just
> about everything running on it, TSM, Oracle, SQL Server, File servers,
> other app servers, etc.
>
> I would say that if your enterprise storage can't handle enterprise load
> levels there's a problem, and it's not with TSM.  I would demand the HP rep
> that sold you the device either refund your money, or show you the
> documentation on load levels the box was covered for.  I'm sure, if you're
> like us, you spent way too much money on this thing for it to be sitting
> Idle.  And, if they didn't properly represent it to you as being unable to
> handle certain loads, they are liable.
>
>
> See Ya'
> Howard Coles Jr.
> John 3:16!
>
> -----Original Message-----
> From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf 
> Of
> Christian Svensson
> Sent: Wednesday, December 07, 2011 8:39 AM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: [ADSM-L] SV: Question concerning large filepools on HP EVA
>
> Hi Daniel,
> I have a HP EVA user that are leaving HP because of similar issues, but
> they have seen it in a different situation. (outside TSM).
>
> About your file class issue. Have you run a q content on does volumes and
> verify that does are really empty so it doesn't contain a pointer.
> Also check if you haven't used all scratch volumes in your file pool so it
> can't reclaim does volumes.
>
>
> Best Regards
> Christian Svensson
>
> Cell: +46-70-325 1577
> E-mail: Christian.Svensson AT cristie DOT se
> Supported Platform for CPU2TSM::
> http://www.cristie.se/cpu2tsm-supported-platforms
>
> ________________________________________
> Från: Daniel Sparrman [daniel.sparrman AT EXIST DOT SE]
> Skickat: den 7 december 2011 14:55
> Till: ADSM-L AT VM.MARIST DOT EDU
> Ämne: Question concerning large filepools on HP EVA
>
> Hi
>
> We have a customer running TSM with a large (+130TB) HP EVA 64000 as a
> large filepool (multi-directory). After a few months, they got problems
> with disks breaking down in the EVA box, ending up having raidarrays in
> almost a constant state of rebuild. After talking to HP, the advice they
> got was to try to reduce the load on the box since it shouldnt be used more
> than 30% of the time during 24 hours.
>
> Initially, the customer used deduplication on the box which probably put
> even more stress on it, but this is now turned off. The problem still
> exists however.
>
> Has anyone else had issues with large diskboxes in combination with file
> device pools? According to HP, this is due to the high amount of I/O that
> TSM produces, but I've seen non-SATA boxes handle alot more I/O than this.
> So the question is, is it because of the use of SATA drives, or is this a
> problem with just this model/box?
>
> The box is equipped with 1TB HP labeled S-ATA disks and the customer has a
> small SAS-based diskbox to handle daily backups and then migrates to the HP
> EVA box. Data is then backed up to a remote LTO-based tape library.
>
> Another problem related to the same pool is that file device volumes that
> has been reclaimed (0.0% usage) is not returned as scratch and deleted, but
> is held within the storage pool as a volume with 0.0% usage. Anyone know of
> any related issues with file device volumes not being deleted?
>
> Customer is at v6.2 on RedHat Enterprise Linux, and we've checked
> permissions on both directories and files of the file device volumes,
> aswell as the TSM activity log, but cannot see any relevant issues.
>
> Best Regards
>
> Daniel Sparrman
>
>
>
> Daniel Sparrman
> Exist i Stockholm AB
> Växel: 08-754 98 00
> Fax: 08-754 97 30
> daniel.sparrman AT exist DOT se
> http://www.existgruppen.se
> Posthusgatan 1 761 30 NORRTÄLJE
> DISCLAIMER: This communication, along with any documents, files or
> attachments, is intended only for the use of the addressee and may contain
> legally privileged and confidential information. If you are not the
> intended recipient, you are hereby notified that any dissemination,
> distribution or copying of any information contained in or attached to this
> communication is strictly prohibited. If you have received this message in
> error, please notify the sender immediately and destroy the original
> communication and its attachments without reading, printing or saving in
> any manner. Please consider the environment before printing this e-mail.
>

<Prev in Thread] Current Thread [Next in Thread>