Re: Large Restore
2000-07-26 10:18:34
Subject: |
Re: Large Restore |
From: |
"Cook, Dwight E" <cookde AT BP DOT COM> |
Date: |
Wed, 26 Jul 2000 09:18:34 -0500 |
We did that with 3.1 & still do it with TSM 3.7
2+ TB of client db file space
compresses down to 560-ish GB transmitted data
currently runs in about 22 hours
Large E10000 client with more CPU's than I'd care to count (runs at about
90+% during backups)
S70 TSM server with a couple CPU's (runs at about 35+% during backups)
fddi loop between the two (that is no router hops)
Slowest part of a restore is unpacking the data and writing it to disk on
the SUN box.
Oh, this is if the data is buffered on a diskpool on the TSM server (where
multiple concurrent sessions can get to it)
if all sessions are waiting on access to a single tape (or limited number of
tapes) nothing will happen faster than you can get the data off the tape
(3590-B1A's going straight out over the network I'd say less than 20GB/hr
compressed client data, maybe more like 10-15GB/hr but that compressed data
should be about 30-45GB/hr client file space...)
hope this helps
later,
Dwight
> ----------
> From: Ewen, Bill S.[SMTP:BSEwen AT WCOM DOT NET]
> Reply To: ADSM: Dist Stor Manager
> Sent: Tuesday, July 25, 2000 10:11 AM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: Large Restore
>
> We are running ADSM 3.1 and we are looking a restoring 400GB of data for a
> test. Anyone out there have any experience with restores this large?
> What
> kind of time frame am I looking at for this restore? Any lessons learned
> that I could incorporate before I start this project?
>
> Thanks
>
> Bill Ewen
>
|
|
|