On Mon, Nov 12, 2012 at 09:52:48AM -0500, Hearns, John wrote: > >> * if PXE server with Infiniband is impossible, then it is OK with a >> gigabyte connection? Or should we go for 16 disks for these 16 clients > >booting is rare and almost trivial, IO-wise. there's no reason you >shouldn't be able to boot several hundred clients over Gb. > > >True. > >I never implemented it, but when I was with Streamline a colleague and I >cooked up an idea >to image large hard drive based clusters using dd and netcat - you can dd >the image of a hard drive, >and use netcat to send that over the network to a fresh node. There is a >multicast option in netcat - ie >you can send out a stream and have multiple listeners. >As I said we never implemented it - not really needed.
http://www.cs.inf.ethz.ch/CoPs/patagonia/dolly.html I used it several years ago, and it worked fine. > >Also as an aside SGI ICE clusters use Bittorrent to ship images around (or at >least did when they used Tempo). As does the Rocks cluster distribution (and derivatives), IIRC. -- Jesse Becker NHGRI Linux support (Digicon Contractor) _______________________________________________ Beowulf mailing list, [email protected] sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
