[Beowulf] Re: Beowulf of bare motherboards
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.caSat Oct 23 13:49:36 PDT 2004
- Previous message: [Beowulf] Re: Beowulf of bare motherboards
- Next message: [Beowulf] Re: Beowulf of bare motherboards
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
> testing. In all cases, I plugged the supply into my Kill-a-Watt, > attached 3 of the above y-cables to the supply, and simply varied the > number of motherboards plugged into those 4 connectors. very interesting, but somewhat hard to interpret. I love my killawatt, too, but unfortunately for this experiment, you need to know how much current the nodes were drawing on each voltage (and the PS specs.) for instance, the TT PS might actually provide 420, but only enough on 3.3 to support a single CPU's VRM, but *lots* of 12V umph. that wouldn't be unreasonable, given the market for "normal" uses - big PS's support machines with lots of disks, or possibly systems that use extra 12V for hot AGP cards, etc. on the other hand, Sparkle is the only one of these PS vendors that I see in OEM settings - TT/MGE/Enermax all seem to be mostly after-market vendors. interpret that how you will ;) regards, mark hahn.
- Previous message: [Beowulf] Re: Beowulf of bare motherboards
- Next message: [Beowulf] Re: Beowulf of bare motherboards
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
