ADSM-L

Performance point of view

2001-02-19 13:02:38
Subject: Performance point of view
From: "Klatt, Michael" <KlattM AT EUROPE.STORTEK DOT COM>
Date: Mon, 19 Feb 2001 17:59:15 -0000
Hi all,

I would like to get some help/infos/ideas for better understanding the
performance point of view.

If some clients are backed up into a disk storage pool, the data (files)
from this storage pool will be migrated into the next storage pool (type
sequential for example) if watermarks are reached. During this migration the
files are stored at the end of the chain "storage pool" -- "device class" --
"library" -- "drives" to drives, to all available drives I assume and  isn't
collocation  enabled all files are distibuted over the volumes, but no
single file is spread out over the volumes. This means to me,  that one huge
file would be stored to just one drive, even if  many drives available,
correct ?

As far as I know about TSM it is different in the TDP for SAP, there I can
stream to different tapes, but what about TDP for Oracle and how can I
manage to get a huge file saved as fast as possible?

many thanks for your help


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