ADSM-L

Re: Collocation issues

1995-03-17 11:40:23
Subject: Re: Collocation issues
From: Paul Zarnowski <VKM AT CORNELLC.CIT.CORNELL DOT EDU>
Date: Fri, 17 Mar 1995 11:40:23 EST
On Fri, 17 Mar 1995 02:50:23 CST Keith A. Crabb said:
>I don't use scratch tapes, I define all the tapes in my tape pools.  What
>I can't seem to figure out is what happens if I have collation set on for
>a tape pool and then define additional tapes to that pool.  Is the server
>immediately going to start up migration processes to move data from tapes
>that have more than one node on them to the new empty tapes, in a effort
>to achieve maximum collation? or is it going to operate strictly off the
>reclamation thresold set for the tape pool?

If you add tapes, the server won't immediately start up migration processes.
The next time a migration process is triggered (because your disk storage
pool started filling up), or the next time a reclaim process is triggered,
ADSM will move filespaces off of disk to a tape (1) preferrably that already
has data from that node on it, or (2) to another tape which does not already
have data from that node on it.  In case (2), I believe ADSM will pick an
empty tape, if one is available.

I should have said in my earlier reply to you that we are using 8mm tapes.
It occurs to me that the impact of using collocation will differ
significantly between 8mm (large, slow) tapes, and 3490 (smaller, faster)
tapes.  With 8mm tapes, for example, ADSM can easily fit data from many
nodes onto one tape.  With 3490 tapes, on the other hand, far fewer nodes
will fit onto one tape.

..Paul
<Prev in Thread] Current Thread [Next in Thread>