Big BLU Sper Computer

[bill at horne.net (Bill Horne)]

----- Original Message ----- 
From: "John Chambers" <jc at trillian.mit.edu>
[snip]

> OTOH, I'm not at all sure of the economics of such things.   I  don't
> really  know  how  to  accurately  calculate  the  cost  of running a
> machine.  Yeah, the cost per KW-hour  is  just  a  number,  but  that
> doesn't make it easy to figure out the actual cost.
>
> Maybe I should get one of those gadgets that gives long-term data  on
> how  much  power  an appliance is using.  Then I could figure out how
> much my current boxes actually cost to run.  I wonder which of  these
> meters  gives  useful  data?  There seems to be an order-of-magnitude
> range of prices, with somewhat vague descriptions of what they do.

John,

Here's the "upper limit" calculation. Keep in mind that it's a "worst case"
figure.

Take the power supply ratings from each machine, and add them together.
Then, divide by 1,000 to get kilowatts. Then, multiply by the number of
hours in a month (e.g., 720) to get kilowatt-hours. Finally, multiply kwh by
your provider's rate.

For example:

10 machines @ 250 watts each =  2500 watts
2500 watts divided by 1,000 = 2.5 kw
2.5 kw multiplied by 720 hours = 1800 kwh
1800 kwh multiplied by $0.06829/hwh = ~ $123 dollars/month.

That's the WORST CASE figure, and assumes you have the same rate I do. YMMV,
and obviously will.

Bill