We just put an Isilon cluster in to production a couple of months ago and so far, I've never had a storage system that came as close to ' just working'. Two thumbs up so far. It's fast and easy to manage and so far has exceeded its specs. We've tried adding and removing shelves and drives while it was under some reasonably heavy benchmarking loads and, well, out just worked. Very impressed so far. We have a 6 node cluster with around 350 tb of data on it, doing around 15-25k iops.
Nicholas On Jun 15, 2011 4:02 PM, "Derek J. Balling" <[email protected]> wrote: > > On Jun 15, 2011, at 2:20 PM, Ryan Dorman wrote: > >> A vote for NetApp ... It just works, > > So I came from a NetApp shop prior to my current gig, and I'm a little curious... > > When you want to upgrade your shelves because, y'know, they're old and slow and running out of warranty, and newer faster better shelves are available now... and you've got live volumes mounted on production systems from those old shelves. How do you do that migration in a way that could be described as "just works"? > > In the LeftHand and Isilon models, it's easy, I add some new shelves into the cluster and remove the old ones, the data-restripes - live - onto the new shelves, and my volumes are just suddenly running on the new shelves without them having any downtime whatsoever. > > My experience in the NetApp world was that this sort of thing is a royal PITA, and can't be done without downtime, and involves copying volumes from one place to another, etc., etc., etc. > > I'm not saying that to be argumentative, but my experience is that "it just works" is something that I'd never use to describe the NetApp experience. It has plenty of advantages, make no mistake, but I'm not sure that that is one of them. :-) > > Cheers, > D > >
_______________________________________________ Tech mailing list [email protected] https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
