On Wed, 28 Aug 2002, Nelson Caparroso wrote:

> Environment: Networker 6.1.1, Shared Library, 18 DLT7000 drives -DDS
> via SmartMedia. Server Parallelism is 42. On a particular Server that
> is a storage node, I have 18 x 70GB superfast mount points that need
> to be backed up in the shortest amount of time possible. Of course the
> fastest way to do this is to stream one mount point (saveset) to a
> drive so all 18 drives are engaged all at once - and I tried using
> target sessions set to 1. However, this scheme only work sometimes...
> like if Networker is taking a long time to figure out what tape(s) to
> load it would still insist in multiplexing save streams to the
> drives... Any other "recipe"  or solution to this problem?

One possibility is to set up 18 duplicate client entries with a separate
tape pool and backup group assigned to each entry. In each entry, specify
a unique saveset path to back up. This is guaranteed to spool data to all
18 tape drives, assuming you schedule all 18 groups concurrently. It will
also greatly improve on restore speed. The down side is you need to devote
tapes to each of the 18 pool AND, most importantly, you need to have the
saveset's files organized in such a way that you can do this reasonably
easily and accurately (you wouldn't want to inadvertantly exclude a
directory from backups).

--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=