Hi All,

Netware restores suck big time when it comes to throughput!
The TSA/SMDR modules are single threaded (as far as CPU's
are concerned - even in Netware 6).
The best we could pusk the Netware client is 4.1M/Bytes per
second - which is faster than what you are seeing but still
an issue when you have 100's of gigs of data to restore.
Novell is aware of the issues but so far have not responded
to a workaround or even an indication of a timeline for
making the TSA multiprocessor aware.
The fact that we use LTO probably helps to achieve a higher
throughput than you.

Regards
Christo

-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU]On Behalf Of
Gallerson,Charles,GLENDALE,Information Technology
Sent: Friday, August 30, 2002 5:29 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Restore Performance.
Importance: High


Mark,

  Does TSM 5.1.0.0 provide MOVE NODEDATA and multithreaded restores

-----Original Message-----
From: Mark Stapleton [mailto:stapleto AT BERBEE DOT COM]
Sent: Friday, August 30, 2002 5:13 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Restore Performance.


From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU]On Behalf Of
Cahill, Ricky
> Doing restores from the collocated primay pools I'm getting 10gb an hour.
> From a non collocated pool (offsite) I'm getting about 5-7gb an hour.
>
> Has anyone got any figures on restores using DLT7000's as we're going
> through the process of doing DR and having to
> restore 7 Netware servers with 1.4tb at the above speeds is going to be
> painful.
>
> PS...ANy hints and tips on doing large restores and how to speed them up
> greatly apreciated as well ;)

You're fighting an uphill battle on two different fronts. DLTs are
relatively slow tapes in terms of throughput, and they are devilishly slow
on mounts and dismounts. *AND* NetWare's TSA module (which waves the baton
for backups and restores) has never been known for promoting fast backups or
restores. (What level of NetWare and TSA/SMDR modules are you at?)

Suggestions:
1. Don't collocate your offsite pools--just your primary pools. Collocated
offsite pools don't buy you much, because your chances of needing them as
small.

2. Upgrade to server version 5.1.1. This will allow you to run MOVE NODEDATA
on your non-collocated pools prior to disaster testing. You can run it by
client, or by client filespace. Version 5.1.1 will also allow for
multithreaded restores.

3. Do the LTO upgrade. Greater capacity (fewer tapes in collocated storage
pools), much faster mounts, and faster throughput are all benefits of such a
move.

4. Use the DIRMC option. Sending directory structures to a disk-based pool
makes restore somewhat faster. (How to do so is a lengthy discussion; Read
The Fine Manual.)

--
Mark Stapleton (stapleton AT berbee DOT com)
Certified TSM consultant
Certified AIX system engineer
MCSE

______________________________________________
"The information contained in this communication is confidential and
may be legally privileged.  It is intended solely for the use of the
individual or entity to whom it is addressed and others authorised to
receive it.  If you are not the intended recipient you are hereby
notified that any disclosure, copying, distribution or taking action
in reliance of the contents of this information is strictly prohibited
and may be unlawful.  Absa is liable neither for the proper, complete
transmission of the information contained in this communication, nor
for any delay in itsreceipt, nor for the assurance that it is
virus-free."