Hi Sam,
I went thru this about two years ago, moving db volume from raw volumes
on Solaris to
a netapp. My experience was that it got faster as the number of db
volumes decreased.
The decreased because the old volumes were 6GB and the netapp volumes
were 20GB.
What I this the problem is, is that TSM updates every dbvolume and log
volume every time
4MB is moved. You know that 1 MB reserved at the start of each volume?
I vaguely remember
hearing that every db volumes knows about every other volume and the
info is in the
1MB reserved.
How many db volumes to you have?
Here are some stats I gathered at the time to report to IBM. They show
the
megs/minute going up as the number of db volumes decreases. If you are
in this situation
there is nothing to do but push through it.
thanks,
Bill Colwell
Draper Lab
------------ stats -------------
minutes dbv size bytes MB/min Comments
1021 5,658,116,096 5.29 68 db volumes
1114 6,446,645,248 5.52
1024 6,446,645,248 6.00
898 6,446,645,248 6.85
913 6,446,645,248 6.73
843 6,446,645,248 7.29
837 6,446,645,248 7.35
853 6,446,645,248 7.21
819 6,446,645,248 7.51
769 6,446,645,248 7.99
722 6,446,645,248 8.52
623 6,446,645,248 9.87 44 db volumes
683 6,446,645,248 9.00
583 6,446,645,248 10.55
589 6,446,645,248 10.44
536 6,446,645,248 11.47
492 6,446,645,248 12.50
467 6,446,645,248 13.16
375 6,446,645,248 16.39
366 6,446,645,248 16.80 28 db volumes
326 6,446,645,248 18.86
293 6,446,645,248 20.98
319 6,446,645,248 19.27
231 6,446,645,248 26.61 21 db volumes
187 6,446,645,248 32.88
335 12,096,372,736 34.44
299 12,096,372,736 38.58
199 12,096,372,736 57.97
168 12,096,372,736 68.67 14 db volumes
-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Sam Sheppard
Sent: Wednesday, December 17, 2008 10:59 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: TSM database on NETAPP
TSM Server 5.5.1 on Solaris10, Sun V240.
I need to relocate my database volumes and have been assigned a LUN on
one of our Netapp FAS6030 devices. I allocated 6 new volumes and then
deleted one of the old volumes (5GB) and the move process started and is
running at about 600MB/hour. I'm not real familiar with the performance
characteristics of the Netapp box, but am assured by the Unix/Netapp
guys that it's a great performer (it's FC-connected).
This seems way slow to me and I haven't seen this poor performance
doing the same kind of operation on my ESS. Anyone have any experience
with the TSM database on one of these devices? I know the original
network-attached storage had write performance problems, but this seems
ridiculous.
TIA
Sam Sheppard
San Diego Data Processing Corp.
(858-581-9668)
|
