On Mon, Jun 18, 2007 at 08:57:29AM -0700, Darren Dunham wrote: > > I think it's mainly to keep you from doing something silly without > meaning to.
That's certainly a good reason. :) > If you have the same type and columns, then you have the same > availability expectations. If instead you take a 5 disk raidz and add a > 2 disk stripe, you're dramatically changing the availability > expectations. So the force is required. Certainly, but adding a mirror doesn't degrade the availability expectations, so it's a fairly safe addition. > It's going to try to stripe across them without regard for performance > today. So one especially slow vdev (maybe a single disk?) could slow > the overall performance for a pool that has other faster components. Ok, so performance will be based on the slowest vdev for the most part. I can live with that. ;) I'm sure this is far from simple, but it would be interesting to see zfs able to figure out the relative performance characteristics of each of its vdevs and spread the load accordingly. :) -brian -- "Perl can be fast and elegant as much as J2EE can be fast and elegant. In the hands of a skilled artisan, it can and does happen; it's just that most of the shit out there is built by people who'd be better suited to making sure that my burger is cooked thoroughly." -- Jonathan Patschke _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
