I work with Ed so I have to be careful here or he'll start throwing sharp
objects over the wall at me- but I don't agree straight-up with everything he's
said.
The Puredisk pricing model does make sense in some environments: our initial
purchase of 2 TB covered 15 offices- that's pretty good math in that scenario
(yes- I know that having servers at that scale doesn't make sense, but not my
call) and compares very well to buying tape drives for all those locations or
copying without dedupe. When it gets ugly is when you get into sites that have
multiple TB - the numbers change drastically.
As for the idea that it should be based on post-dedupe data size: good luck
with that! How exactly you can calculate that in advance with any confidence
has eluded me- we have everything from 20:1 to 1:1 depending on the
site/application. You'll end up overbuying or underbuying and getting pressure
from the bean-counters to "improve your assumptions." At least with the
pre-dedupe prices you can point at the disk and say "if you want to pay less-
shrink that." It actually works sometimes!
You are also assuming that Symantec would charge the same price/TB if they
changed the model to post-dedupe capacity vs pre-dedupe. Not likely- any
change is likely to be "revenue neutral" at least in the short term.
Tom Burrell
P.S: Ed -what's this "we" thing on WHS? You're on your own there, pal! ;-)
ewilts wrote:
> On Thu, Apr 23, 2009 at 10:24 PM, Tristan Ball <Tristan.Ball < at >
> leica-microsystems.com (Tristan.Ball < at > leica-microsystems.com)> wrote:
>
> >
> > I?d second both of these points, but especially the pricing one. When we
> > looked at it, pricing was based on the volume of data pre-deduplication,
> > and bought in blocks of n Gigabytes or Terabytes.
> > 1) From a business point of view the one of the big value point of
> > dedupe is to reduce the storage/transfer costs of backup. But their pricing
> > model means that rather than spending that money on infrastructure (local
> > tape, faster links, whatever), we pay that money to veritas/symantec. OK,
> > fine, it?s probably less money over all, but even so to me it means that
> > Symantec is getting the lions share of that value from the software, not
> > me. I actually think this will change somewhat as dedupe becomes commodity.
> >
> >
>
> Becomes a commodity? In some ways, it already is. For $100 I can buy a
> Windows Home Server license and de-dupe the backups for an infinite number of
> TB from 10 PCs in my house. I've got about a half-terabyte at home being
> de-duped and backed up to a $500 server. If I had a similarly sized PureDisk
> environment, I'd easily pay 10 times that.
>
> Seriously, we've considered WHS for some small offices because the PureDisk
> license costs are ridiculous.
>
> I absolute HATE per-TB licensing. For a lot of what we do, this model breaks
> down. We're not that large of a company, but we have a lot of storage.
> Vendors make the silly assumption that the more storage you have, the richer
> you must be. It's like charging a $100 to do an oil change on a Rolls Royce.
>
> I just wish that Symantec would sell me the software and let me determine how
> many TB I can throw behind it.
>
> I don't pay per-TB licenses for NetBackup and I don't want to pay them for
> PureDisk. The only reason we do today is because Symantec's competitors
> haven't seen the light yet and commoditized it yet like Microsoft has with
> Windows Home Server. As soon as they do - or if Microsoft licenses that
> backup component separately in an enterprise, Symantec will lose our business.
>
>
>
>
> > .../Ed
> >
> > Ed Wilts, RHCE, BCFP, BCSD, SCSP, SCSE
> > ewilts < at > ewilts.org (ewilts < at > ewilts.org)
>
+----------------------------------------------------------------------
|This was sent by tburrell_mn AT yahoo DOT com via Backup Central.
|Forward SPAM to abuse AT backupcentral DOT com.
+----------------------------------------------------------------------
_______________________________________________
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
|
