Then we need to reevaluate what the purpose of Solaris is, because there's still no server variant of OpenSolaris, there's just too much fat on it still, even with Indiana. The target is obviously users at first, not sure which ones but calling this experiment more than a tinker toy at this point is probably better than claiming it's enterprise grade since I've seen a fair share of dumb things done so far with the project regarding these sort of choices.
Since OpenSolaris is not encumbered by seat and customization restrictions, a spin for large companies and individuals who want parity of their spins across their work and personal lives should be made available so these sort of penalties do not affect them. At this point no one really knows what the intention of OpenSolaris in general is, let alone the target demographic, hardware or environment. James On Oct 15, 2008, at 11:22 AM, Bob Friesenhahn wrote: > On Wed, 15 Oct 2008, James Cornell wrote: > >> Meta tagging on only newly created files is a better way, since the > > If the computer is attached to 300TB of file data, how do you > propose that this "tracker" know about the new files without > spending weeks rummaging through the filesystems, consuming lots of > CPU, network bandwidth, storage IOPS, and power? Now imagine the > effect if 10,000 computers running "tracker" are attached to that > same 300TB of file data. > > This sort of facility has no business being included by default in > an enterprise OS. It is like handing a grenade to a child with the > pin already pulled and asking him not to let go. > > The idea of a content addressable filesystem is appealing but it > does not scale. > > Bob > ====================================== > Bob Friesenhahn > bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ > GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ >
