Our storage pool is colloated,,,there are only 3 volumes used for this
server's backed up data. Also, there were no tape drive contention
problmes when I ran the restore...that's why this is so puzzling.
Debbie
Troy Frank <Troy.Frank AT UWMF.WISC DOT EDU>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
08/03/2005 05:27 PM
Please respond to "ADSM: Dist Stor Manager"
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: Re: [ADSM-L] Technote 1200328
It looked to me like it could just be a case of lots of small files spread
out across lots of tapes. Average file size was ~200k, with ~27,500
files. With that many small-ish files, if they were spread out across
many tapes, you'd spend most of that 56min waiting for tape mounts/seeks.
Short of collocation, not much would save you in a situation like that. If
all these files were basically on 1or2 tapes, this theory falls apart,
however.
Troy Frank
Network Services
University of Wisconsin Medical Foundation
608.829.5384
>>> Deborah_Bassler AT PAPAJOHNS DOT COM 8/3/2005 3:19:53 PM >>>
Oops, I meant to include that in the email. The bottom of this doc shows
the transfer rates..
Lawrence Clark <Larry_Clark AT THRUWAY.STATE.NY DOT US>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
08/03/2005 04:10 PM
Please respond to "ADSM: Dist Stor Manager"
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: Re: [ADSM-L] Technote 1200328
What did the tranfer rate show as?
1024 x 6.24 = 6389MB (megabyte)
Network is usually in Megabit, yes?
6389MB x 8 = 51112 (megabit)
>>> Deborah_Bassler AT PAPAJOHNS DOT COM 08/03/2005 3:49:17 PM >>>
This doc offers alot of information about improving performance. I'm
especially interested in this because it took 56 minutes to restore
6.24G
of data, from Novell server to Novell server, over a 100MB pipe.
Our TSM version is 5.1.1 ( I know,,,we need to upgrade)...and the
client
version is 5.2.
In the dsmserv.opt file the MIRRORWRITE DB = SEQUENTIAL. According to
this
doc, we'll get better performance is we change MIRRORWRITE DB to
PARALLEL.
I thought I would do this then add the DBPAGESHADOW = YES parameter.
(the
MIRRORWRITE LOG = PARALLEL)
My plan is to make small changes to see if there is an impact, positive
or
negative. We have 2G of virtual memory, so I changed the bufpoolsize
from
262144 to 524288 and thought I'd make the MIRRORWRITE DB change also.
Has anyone made these changes and seen any performance
improvements/degredations? Any experiences or advice is welcome.....
Thanks for any input,
Debbie
Confidentiality Notice follows:
The information in this message (and the documents attached to it, if any)
is confidential and may be legally privileged. It is intended solely for
the addressee. Access to this message by anyone else is unauthorized. If
you are not the intended recipient, any disclosure, copying, distribution
or any action taken, or omitted to be taken in reliance on it is
prohibited and may be unlawful. If you have received this message in
error, please delete all electronic copies of this message (and the
documents attached to it, if any), destroy any hard copies you may have
created and notify me immediately by replying to this email. Thank you.
|
