[Beowulf] Academic sites: who pays for the electricity?

Robert G. Brown rgb at phy.duke.edu
Thu Feb 17 20:31:33 PST 2005

On Wed, 16 Feb 2005, Vincent Diepeveen wrote:

> >  1000  (nodes) *
> >     2  (cpus/node) *
> >     .1 (kilowatts/cpu) *
> >     .1 (dollars/kilowatt-hour) *
> >  365   (days /year) *
> >   24   (hours/day) =
> >-----------------------
> >  175200 dollars/year
> Complete academic nonsense calculation. If you use quite some electricity
> the electricity gets up to factor 20-40 cheaper. Getting a factor 10
> reduction in usage bill is pretty easy if you negotiate properly.
> However you must avoid starting machines at peaktimes. Big fines get given
> for that. So it's cheaper to let them run 24 hours a day than to start them
> in the morning after say 7 AM (depending upon local habits).
> Please note that nothing beats the price of nuclear power 
> (as a member of the high voltage power forum i do not have an opinion on
> that). 
> Electricity production costs of nuclear power are hundreds of times cheaper
> than producing it with oil, oil produces it roughly for 5 dollar cent a
> kilowatt (if memory serves me well). Coals have a CO2 problem for nations
> which are in Kyoto agreement (USA isn't), but also is nearly as cheap as
> nuclear power. 
> So the actual price they deliver huge power for to big institutes is a very
> easy negotiation to get it factors down.

Actually, in spite of the fact that Duke (partly) owns its own power
company, I don't think that they get all that much of a discount.  I'm
also quite certain the the nuclear plant about 20 miles from here
doesn't sell its electricity to customers across the county line (it's
not a Duke Power plant but rather CP&L IIRC) for any less than Duke and
Durham get it.

There is a nifty map here:


that shows the average electricity costs throughout the USA as of 2001
-- they are almost certainly a half-cent/KW-hr higher across the board
if not a whole cent higher due to the war and oil price boosts.  Note
that they range from $0.04 and change in major coal states to $0.11 and
change per KW-hr (where I'm sure a chunk of the difference is taxes in
different states).

My recollection from discussions at Duke is that Duke pays around
$0.06/KW-hr, not a huge discount over what I pay at home.  Maybe major
industrial consumers of electricity do better in states like Michigan,
but I don't think that there is enough margin even in the coal states to
drop prices to $0.01/KW-hr after paying for the fuel itself for anything
but nuclear.

Where David lives, in CA, electricity is about as expensive as it is
anywhere.  At a guess currently over ten cents/KW-hr, probably even for
Universities.  David?


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 mailing list