Power consumption -- real vs actual
jal at mdacorporation.com
Tue Apr 6 16:55:21 CDT 2010
> To turn this around a bit, what question are you wanting to answer with
> the power consumption numbers?
> Are you trying to size cooling systems? Size UPS or PDUs? Evaluating
> operating costs?
yes, yes, yes, no.
For a particular customer I have a problem with weight, power, and size. (I guess that's three problems!) Mostly power, as it is diesel generated and we have to size the generator. Keeping total power under the genset limit is the goal, otherwise we have a big jump in costs to a larger genset.
> I've found the tools are dell.com/calc to be pretty good at reproducing
> what we measure using a power meter. One caveat is that they give the
> same power for all bins of a XEON CPU --- same for all L, E, X clock
> speeds, which isn't really the case. This results in overestimates for
> some CPUs.
A little overestimation (e.g. 30watts in 450, or 7%) I can handle. We can do margining after that.
What I'm seeing the the dell.com/calc to be sort of OK for is enterprise level stuff (not all equipment is handled equally, e.g. there is nothing in there for workstations), it is a little uneven in U.S. vs metric measures, etc. And no guidance on what actual power is required vs VA ratings vs power distribution vs cooling -- hence my topic title.
Data on UPS would be ideal. A UPS takes in, say, 2900VA, converts it to 2650 VA on output and consumes the difference, in watts, (250 watts) in losses emitted as heat. dell.com/calc does not seem to handle this at all, although it allows to show one in the physical configuration.
More information about the Linux-PowerEdge