ADSM-L

Re: backup stgpool -- everything offsite

2005-09-09 19:51:06
Subject: Re: backup stgpool -- everything offsite
From: "Allen S. Rout" <asr AT UFL DOT EDU>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Fri, 9 Sep 2005 19:50:49 -0400
==> On Fri, 9 Sep 2005 15:15:12 +1000, Steven Harris <steve AT STEVENHARRIS DOT 
INFO> said:


> Excuse the obvious question, but what happens if there is a disaster at the
> *remote* site?  Whilst you'll still have your current disk copies of
> everything, you'll lose all your historical stuff, and those "never to be
> deleted" backups that managers are so attached to.  You'll also have no
> capacity to perform day to day backups without the remote site, should it be
> out for an extended period.

I'm sorry, I've evidently been unclear; let me try again.

Right now, I've got a primary stgpool heirarchy, an onsite copypool heirarchy,
and an offsite copypool heirarchy, physical transport.

In the planned arrangement, there will be a primary stgpool heirarchy (onsite)
and a set of copypools, offsite via electronic transport.

So the remote disaster means that I've only got the primary copy of the data,
and am exposed to local site disaster until such time as I can generate an
additional copy.

Is that more clear?


For those data sufficiently important that merely two copies is inadequate,
I'm still making an onsite copy, and yet another offsite copy through another
electronic vaulting arrangement, for a total of four different locations.





> Finally, you are also making your whole strategy dependent on something over
> which you have no control, ie, the reliability and redundancy of the network
> carrier.  I wouldn't trust any of our local ones in the long term, as I'm
> sure some bean-counter will pare their maintenance budget past the minumum,
> and network failures *will* result.

Certainly an important evaluation to make.  In this case, I've got a situation
which makes me feel confident of the intervening network.  Our "long haul"
carrier is the Florida LamdaRail (FLR) .  Most of the engineering staff of FLR
work one floor down from me, and their director reports to my director :) Oh,
and their backups are on my TSM server.  heh.

Doesn't mean that there will be zero downtime, but it does mean that they care
about my service.



- Allen S. Rout

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