> From: Luke S Crawford [mailto:[email protected]]
> 
> I've got a kill-a-watt (and several remote PDUs, which is what
> I normally use, but as we're doing this on pre-production equipment,
> I can use the kill-a-watt to make the test the same as yours.)

I'd be interested to see if the kill-a-watt measures the same as your pdu.

And it's important to have both of the power supplies of the server running
through the kill-a-watt, so I use a 3-way splitter on the kill-a-watt.


> How many disks do you have in the server when you measure?  or do
> you want to do the measurement without disks?

I'm not normally measuring for the sake of figuring out how much the
*processor* consumes.  I normally want to know how much the server draws, so
I measure the server in whatever configuration the server is going to be
used.

This means the result of comparison of whatever measurements we may each
have is ... perhaps ... less than extremely valuable.

I guess the thing that would be really useful would be to know the actual
power draw of any given system, in any given configuration.  So you could
actually use that information to help guide your purchasing decisions.
Unfortunately, that's too much to ask of vendors.  And it seems unrealistic
to turn this into a community project, website, or wiki.

I guess I will say this:  Particularly for the units you've pointed at on
supermicro, if you have power measurements for those systems, it could
affect my purchasing decision, to use supermicro or not, and specifically to
use that unit, or not.


> > I use a kill-a-watt.  When the server is new, I write a 3-line python
> > script, which is an infinite loop of random number generation.  I
> launch as
> 
> What's the three lines?   I want to use the same three lines you do.

#!/usr/bin/python
import random
while True:
  random.random()

./random.py &
./random.py &
./random.py &
./random.py &
./random.py &
./random.py &
...


> I'm smaller than you are, apparently.  The cost of cooling is rolled
> into
> the cost of my power.  

Same here.  Except, that your cooling system needs to be appropriately
sized.  And without power measurements, you can't conclusively say if you
need a 10,000 btu air conditioner from home depot, versus half a dozen units
on the roof of the building.


> Also, as far as I can tell, adding a UPS, when
> you are already at a datacenter with a good UPS and generator doesn't
> add anything besides something else to fail

I'm at a datacenter that just provides wall power.  Add your own UPS.  No
generator.


> I believe that heavy seeks vary the power used by the disk by a
> good bit.  (sequential transfers, not so much.  It's waggling the read-
> write
> head about, as far as I can tell, that varies the power draw.)

I haven't been able to see any difference, whether I'm doing a "while true"
loop around dump or tar, versus dd.  But it's certainly possible.

_______________________________________________
Tech mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to