ADSM-L

Re: [ADSM-L] big striped TSM volume performance

2009-05-19 06:04:16
Subject: Re: [ADSM-L] big striped TSM volume performance
From: David McClelland <david.mcclelland AT NETWORKC.CO DOT UK>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Tue, 19 May 2009 11:02:37 +0100
Take a look at this link to a presentation given by one of the TSM
Developers at the Oxford Symposium a couple of years back which will answer
one or two of your questions and give some IBM recommendations for best
performance on different disk subsystems and layouts >>>
http://tinyurl.com/qv35v8

Part of the answer below may also depend upon what you mean by 'big TSM
volume' - a random access DISK volume, or a sequential access FILE volume?

HTH,

/David Mc
London



-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Mehdi Salehi
Sent: 19 May 2009 10:12
To: ADSM-L AT VM.MARIST DOT EDU
Subject: [ADSM-L] big striped TSM volume performance

Hi,
Which one performs better in TSM for AIX:
1- using one big TSM volume on a big striped filesystem
2- using multiple smaller TSM volumes each on a separate filesystem (each
filesystem resides on a separate hdisk)
(in either case, disks are RAID protected in disk subsystem layer)

I assumed that TSM does not lock a volume while a client is sending backup
data to it such that other clients can read/write from/to this volume
simultaneously. Correct me if I am mistaken, please.

Thanks so much,
Mehdi Salehi

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