If your target server storage pool is a backup pool, then the reclaim  on
the virtual volumes will be built from the primary pool on the source server
and sent over the network in only one direction (although sometimes, 1% of
the time, it does pull some data across the network in both directions.)

From: "Norback, Jan" <Jan.Norback AT NL.ORIGIN-IT DOT COM>
Reply-To: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Server to Server reclamaiton vs move data
Date: Tue, 1 Aug 2000 16:54:34 +0200

Jack,
I you work server-to-server all your reclaims will go from the remote to
the
local and back to the remote again thus passing your network twice! Same
thing for move data.

This, hopefully, should be nicer in a SAN environment but I have not seen
that yet.
Regards,
Jan Norback

Origin-IT (MS/DS/OSS Unix)
VA-173, PO-box 218, 5600 MD Eindhoven, The Netherlands
Groenewoudseweg 1 5621 BA Eindhoven, The Netherlands
Email: Jan.Norback AT nl.origin-it DOT com
Tel: +31 (0)40-2780289
Mobile: +31 (0)6-25233507
Fax     : +31 40 2783962


-----Original Message-----
From: Palmadesso Jack [mailto:Jack.Palmadesso AT SWPC.SIEMENS DOT COM]
Sent: Tuesday, August 01, 2000 4:29 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Server to Server reclamaiton vs move data


Boy sure is quiet in here lately.  Maybe this can stir things up a bit ;)

Correct me if I am wrong:

I have 2 servers at remote sites that perform server to server db backups
and copy storage pools over our WAN.  Reclamation usually takes a really
long time to complete.  Especailly as I approach lower reclamation values.
If I set reclamation for the server to server pool at 80 then I have to
send
at most 20% of each tape down over the WAN.  DLT 4000s have 20/40
compression so I could send up to 5/10 gigs of data down the pipe for just
for one volume!!  That could take all day over the WAN!

I am considering using "move data" instead.  Again correct me if I am wrong
please.

When I issue a move data on a server to server volume that volume should be
mounted in the remote library as well as another filling volume from that
storage pool or just use a new scratch volume.  Since these two volumes are
mounted in the same library the amount of data that needs to be sent over
the WAN is now minimal (I'm not sure about this)  In this case the data
would be moved much quicker freeing up my scratch volumes.  Also
communication over the WAN is now minimized because no files are being sent
only ADSM DB communication is going on.  I know I would need at least 2
mount points available to the storage pool.  Can anybody enlighten me on
this.

Thanks for any help as always,

Jack

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com