Save Energy: Promoting Hibernation for Computer Users

[Loren Booda]

I believe that if computer users knew how to hibernate their computers, they would save about 100,000,000,000 kWh yearly in the United States alone. How can the Web spread the word?

 

[Hurkyl]

I believe that if computer users knew how to hibernate their computers, they would save about 100,000,000,000 kWh yearly in the United States alone. How can the Web spread the word?

Is that really accurate? Is it really significant?

 

[waht]

The problem is that it's unbearable.

 

[Loren Booda]

Is that really accurate? Is it really significant?

It represents .3% of total U.S. consumption (Wikipedia). Our country is becoming more aware that many small savings add to a large saving. Energy may well be the currency of the future (Mad Max?). Maybe California's outages would be less brown with home computers hibernating half the time (in part, while the human is at work and air conditioning peaks).

 

[Hurkyl]

It represents .3% of total U.S. consumption (Wikipedia).

Where on wikipedia?



Anyways... saying that this would save 1000000000000 kWh yearly is an example of what not to do. Does that number really hold any meaning for you? It doesn't for me, nor I imagine the average American. You might be able to sucker some people with 'big number blindness', it is a deceitful tactic and I thoroughly condemn it. (And, of course, it can backfire when presented to someone practicing critical thinking)

It represents .3% of total U.S. consumption (Wikipedia).

This is a better figure (assuming its accurate), but it still needs a good citation. In my opinion, it needs more elaboration. There are a lot of things this could mean. For example, is this just home computers, or does it include computers at the workplace? How much hibernation are we talking about? 24-7? 8 hours a night? You need a usage scenario that fairly clearly specifies what's being measured, and ideally, the calculation of the figure needs to be put in a footnote / addendum, or at the very least, easy to find in the citation.


However, the fact that this is such a small number would suggest that, maybe, it's not worth it. How does the energy I waste by not putting my computer into hibernate mode compare with, say, leaving a light bulb on? How many cents an hour does it cost me? Is this particular method of saving energy really worth my effort as compared to other things I could be focusing on?

Energy may well be the currency of the future (Mad Max?).

No. Don't do this. You appear to be attempting to associate current energy usage with a post-apoctalyptic scenario. This is another deceitful tactic -- you are trying to sway opinions based upon this juxtaposition rather than the strength of your argument.

Maybe California's outages would be less brown with home computers hibernating half the time

Maybe, maybe not. Seek actual figures, and use them. (And if you cannot find any, then leave out such idle speculation)



P.S. I added an extra zero when I quoted that energy savings figure. Or maybe I took one out, I can't remember, and can't tell by looking at it. Did you notice? That is a big clue that this number is meaningless.

 

[vociferous]

I hibernate my laptop, which I only use when I am away from my desk. My computer at home is always performing tasks (including VPN gateway), so I have to leave it on all the time, which I am sure sucks up the Kilowatts. It has a 1500 watt power supply, though it probably uses less than half of that when it is idle.

 

[shoehorn]

Energy may well be the currency of the future (Mad Max?).

Preposterous nonsense. Currency will be the currency of the future.

Regardless, it's hard to have a great deal of sympathy for energy woes in America when the "high" gas prices there actually aren't high at all.

Maybe California's outages would be less brown with home computers hibernating half the time (in part, while the human is at work and air conditioning peaks).

Without detailed data to hand, this falls firmly in the "baseless speculation" column. If California really wanted to prevent brownouts a far better place to start would be to invest more heavily in better structured derivatives in the energy markets.

 

[B. Elliott]

I think it could provide considerable savings to many calls centers in the US. At my old job I tried to talk the IT and administration department into hibernating the computers they used out in the call center. They kept every computer (even ones which were unused) running all the time along with every monitor always on.

At any given time there would be approximately 200+ computers which were unused and idling with no power save features on. Multiply that over a one month time and that could add up to considerable savings. Multiply that by say, 100+ call centers and that would be a considerable amount of power being saved.

 

[Loren Booda]

Please peruse ESTIMATING TOTAL POWER CONSUMPTION BY SERVERS IN THE U.S. AND THE WORLD, at http://enterprise.amd.com/Downloads/svrpwrusecompletefinal.pdf". It states "Total power used by servers represented about 0.6% of total U.S. electricity consumption in 2005. When cooling and auxiliary infrastructure are included, that number grows to 1.2%, an amount comparable to that for color televisions. The total power demand in 2005 (including associated infrastructure) is equivalent (in capacity terms) to about five 1000 MW power plants for the U.S. and 14 such plants for the world."

My percentage calculation - more an estimate, (.3%) - was quite close, but my total was way off. I should have compared more sources on total electrical consumption in the U.S.

 

[Hurkyl]

Please peruse ESTIMATING TOTAL POWER CONSUMPTION BY SERVERS IN THE U.S. AND THE WORLD, at http://enterprise.amd.com/Downloads/svrpwrusecompletefinal.pdf".

This appears to have absolutely nothing to do with what we're discussing.

 

[Loren Booda]

I would say that, given the total power consumption of servers (.6% of all electricity consumed in the United States, as related by the aforementioned article) and a rough average for percentage savings of modes like hibernation in those computers, one can estimate a ceiling for overall energy saved in PCs. If this average savings for hibernation is 50%, the article's figure of .6% would seem to imply a savings by personal computers of at most .3% of the total electricity consumed in the United States.

Does anyone else have more relevant information to offer?

 

[Hurkyl]

I would say that, ... one can estimate a ceiling for overall energy saved in PCs.

And you would be wrong -- this article doesn't say anything about the desktop computers people have sitting at home.

(P.S. I'm still waiting for that citation for your wikipedia reference)

 

[Loren Booda]

I am sorry, Hurkyl, I misstated that figure. I had used the data given at http://en.wikipedia.org/wiki/Personal_computer that "More than 500 million PCs were in use in 2002 . . . the United States had received 38.8 percent (394 million) of the [personal] computers shipped." Questionably, I use the figure of each PC as using an average of 30 watts continuous. This might indicate a total PC power consumption in the United States per year of

24(hours/day) x 365 (days) x 30 (watts/PC) x 394,000,000 (PCs) = 1.04 x 10¹⁴ (watt-hours) = 104 TWh

The total energy consumption in the United States is taken from http://en.wikipedia.org/wiki/Energy_use_in_the_United_States" "The United States is the largest energy consumer in terms of total use, using 100 quadrillion BTU (105 exajoules, or 29000 TWh) in 2005."

104 TWh/29000 TWh = .36%, the purported consumption by PCs in terms of total US energy usage.

 

[CRGreathouse]

24(hours/day) x 365 (days) x 30 (watts/PC) x 394,000,000 (PCs) = 1.04 x 10¹⁴ (watt-hours) = 104 TWh

I would expect that most computers are on for less than 8 hours per day. Hibernation could help some of the 24-hour users cut down their usage, and standby could reduce the power use during the time that computers are nominally used, but I certainly don't think 0.3% of the total power could be cut.

Of course 30W per computer seems low to me, so perhaps the errors cancel out.

 

[Greg Bernhardt]

I always set my sleep timer for 10min, seems to work well and doesn't use much power. i can't stand booting up, takes forever.

 

[B. Elliott]

Also take into consideration the monitors power requirement. LCDs are typically lower on the order or 20-60watts depending on size, while older CRTs typically draw anywhere from 60-90watts. Some are even higher than that at around 120watts. The monitors also constantly draw power even when completely switched off (8-10watts?)