DUAL CPU board vs 2 Single CPU boards: bang for buck?
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Mark Hahn hahn at physics.mcmaster.caThu Mar 7 10:31:37 PST 2002
- Previous message: DUAL CPU board vs 2 Single CPU boards: bang for buck?
- Next message: DUAL CPU board vs 2 Single CPU boards: bang for buck?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
besides the blade and 1U approaches, I've been wondering about using small-form-factor uATX cases. these are roughly 3x12", which is pretty close to the same "rack surface area" as 1U, so the density would be comparable to 1U uniprocessors. you can get them on the commodity market (eg AOpen uATX cases), but vendors like Dell and Compaq both have fairly attractive units. the big vendor approach is attractive because of preconfiguration and the 3-year warantee they come with.. prices seem pretty good - starting at around $US 1k for a nice little P4 brick. Compaq, for instance, seems flexible about things like omitting kbd/floppy/disk/video/windows for substantial orders, and their other Evo machines definitely PXE boot. does anyone have horror stories about trying this sort of thing? these are definitely not plain old beige boxes, and for people who simply need >100 CPUs with unexceptional network, possibly no disk, this seems pretty attractive. thanks, mark hahn.
- Previous message: DUAL CPU board vs 2 Single CPU boards: bang for buck?
- Next message: DUAL CPU board vs 2 Single CPU boards: bang for buck?
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
