Measuring energy readings

[JMBates]

Hello all,

 

We want to start measuring how much energy our network is using. We have a very vague kWh usage for the day and night a long with the costs per month but I've got a feeling it won't sure any progress. We are going to install a SAN and get rid of our server sprawl hence the reason to measure our energy usage to get a before and after of installing the SAN to see if it is actually saving energy. Does any one have any solutions on how to measure this?

 

Thanks!

[glennda]

What UPS do you have? you may it gives you the info you want.

[JMBates]

It's an APC UPS but it doesn't have a management card in it to monitor usage unfortunately.

[glennda]

can you still not monitor via serial/usb don't necessarily need a network card.

 

Does it not have serial/usb?

[AngryTechnician]

If all else fails, get a power usage meter like this and plug it between your UPS and the mains. As long as the UPS power draw isn't over 13A (probably isn't, but check the fuse) it will work just fine.

 

Obviously take the reading when the UPS is not recharging!

[JMBates]

If all else fails, get a power usage meter like this and plug it between your UPS and the mains. As long as the UPS power draw isn't over 13A (probably isn't, but check the fuse) it will work just fine.

 

Obviously take the reading when the UPS is not recharging!

 

Thank you for this. I'll look into getting one seeing as it's cheap. Have you got one set up?

[AngryTechnician]

I don't have one on our UPS any more as we have management cards in them now, but I have used one in the past and it seemed to work well. Also good for demonstrating to disbelieving Bursars how efficient Sleep mode is on modern PCs!

[FN-GM]

Just out of curiosity does anyone know about how much it cost per hour to run a pc please?

[Sylv3r]

Just out of curiosity does anyone know about how much it cost per hour to run a pc please?

 

I'm no expert but it will all depending on the PSU etc inside it.

[FN-GM]

I'm no expert but it will all depending on the PSU etc inside it.

 

Well yes, but i am talking about averages really.

[pcstru]

Depends dunnit. I reccon the average PC+TFT draws between 80-120 watts on idle. @10p per kwh (probably more), that's 19-28p per day if they run 24 hours.

 

For a school, buying 100 PC's that are 10wph more efficient than the competition is likely to save ~£160 pa (+) (9 hour day 180 days per year) (a saving that is only likely to increase).

[Arthur]

Just out of curiosity does anyone know about how much it cost per hour to run a pc please?

How Much Does Your PC Cost To Run? « Bit-Tech

[JMBates]

I've bought a plug-in electricity cost monitor myself and will be monitoring how much electricity we are using in one of our server racks over the upcoming weeks. We will then be comparing these figures after we install our SAN to see if there are any improvements. Thanks for the posts guys!

[PeteHopton]

A Good rule of thumb is that 1W left on is £1 per year.

 

So if you've got 100W, on all the time £100 per year, if you've only got it on half the time then it's £50 per year.

 

this works out at about 11p per KWh.

[AngryTechnician]

A Good rule of thumb is that 1W left on is £1 per year.

 

So if you've got 100W, on all the time £100 per year, if you've only got it on half the time then it's £50 per year.

 

this works out at about 11p per KWh.

 

Sorry to be contrary, but in my book a good rule of thumb is to never rely on rules of thumb. I actually ran the numbers for our PC estate a little while ago, and the cost of our machines left running was much less than the sort of figures that get bandied around by people like Gartner and Microsoft. Not only are our electricity prices significantly lower than 11p/kWh, but our hours/day were lower too because we actually have power saving policies defined on all the machines.

 

My advice is always to spend a few quid on a meter and do the measurements for your kit and setup. If you go with the sort of numbers that I see regularly on energy efficiency reports, you could end up with egg on your face when the actual savings come in at much less than what you've promised to your finance department.

Edited September 7, 2012 by AngryTechnician

[SYNACK]

Those cheap meters are not really a good measure as they are generally designed for constant loads, computers with advanced switching mode PSUs are a tad more complex in their energy usage so the basic averaging done by those little plugs can be off by a margin. Fine for a before/after comparison but not for a truly accurate measurement.