ADSM-L

[ADSM-L] Sequential disk storage pool performance

2012-07-06 15:00:03
Subject: [ADSM-L] Sequential disk storage pool performance
From: Thomas Denier <Thomas.Denier AT JEFFERSONHOSPITAL DOT ORG>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Fri, 6 Jul 2012 14:53:06 -0400
We have two TSM 6.2.2.0 TSM servers in different locations running under
zSeries Linux. Both use sequential disk storage pools on IBM XIV disk to
store the primary copies of backup files. The databases and recovery logs
are also on XIV disk.

We are still in the process of migrating workload to the two sites. We
expect to end up a couple of hundred TSM client nodes at each site, with
something in the neighborhood of a terabyte per night of incoming backup
data at each site.

We will shortly be taking delivery of a Hitachi HUS150 disk array for each
location. The initial motivation for these was to provide significantly
cheaper storage for large image repositories. We are considering using
the new arrays for long term storage of TSM backups as well. Two
different approaches to this are under consideration. One approach is to
have incoming backups written to HUS150 disk and then left there. The
other approach is to have incoming backups written to XIV disk and
subsequently migrated to HUS150 disk for long term storage.

Has anyone tried the first approach with an HUS150 or a competing product
with similar performance characteristics?

Thomas Denier,
Thomas Jefferson University Hospital
<Prev in Thread] Current Thread [Next in Thread>
  • [ADSM-L] Sequential disk storage pool performance, Thomas Denier <=