Robert G. Brown
rgb at phy.duke.edu
Tue Jan 14 15:19:42 PST 2003
On Tue, 14 Jan 2003, Michael Stein wrote:
> > High density clusters also carry a fire risk of their very own. One can
> > easily achieve node densities that consume 2 or 3 thousand watts in a
> > rack (and with effort, can maybe double that).
> I've seen a dual 2.4 GHz Xeon Intel 1U machine measured at 250 W (multiple
> "burn*" running). A rack full of these (42 U) really is 10KW.
Has anyone actually filled a rack with nodes this hot and burned this
much power, per node, sustained? Kind of scary, if you have;-)
A joule is dropping a one kilogram hammer on your finger from 10 cm.
So 10KW is a one-metric-ton jackhammer, pounding up and down one meter
once a second, sustained. With all the cooling fans running, it
probably sounds like one as well...:-)
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
More information about the Beowulf