PowerSwitch

We discussed the energy internet use a while ago but has anyone any recent data or links to reports on the topic.

I accidentally got 'volunteered' into writing a piece for a worthy group of people.

The NSA, GCHQ etc probably use a lot of energy storing all our communications data .

google hosted a 'how green is the internet' summit last month:
http://www.google.com/events/howgreenistheinternet2013/

There's various pdfs' including 'direct energy use of internet flows' or something.

Maybe of use, or maybe not. My phone can't open pdfs so unable to check.

Also, nytimes article last sept.
http://www.nytimes.com/2012/09/23/techn ... wanted=all

Thanks Snail.

It's probably important to clarify what you mean by the energy internet.

There is the energy used to run the millions of PCs that wouldn't otherwise be on but for the Internet.

Then there are the server farms that house the servers that provide the power and storage to display pictures of cats for you. These also have huge associated cooling costs.

Then there is the energy required to power the telecommunications infrastructure that the data moves on. The problem here is that modern equipment that allows fast communication is far more power efficient than that old POST equipment it replaced.

Then there is the increased load on the energy generating capacity caused by people charging their Internet aware mobile devices that probably wouldn't exist but for the Internet.

But then you could offset the energy that is no longer expended by people travelling to meetings that are now held virtually or going shopping in town now that DODGY TAX AVOIDERS is available or even carrying billions of small pointless envelopes around now that their contents can be emailed.

JavaScriptDonkey wrote:But then you could offset the energy that is no longer expended by people travelling to meetings that are now held virtually

So why is HS2 needed?

JavaScriptDonkey wrote:
Then there are the server farms that house the servers that provide the power and storage to display pictures of cats for you. These also have huge associated cooling costs.

Not just cats. I've heard tell that some electricity is used for the display of scantily clad ladies.

But I make no judgements.

JohnB wrote:

JavaScriptDonkey wrote:But then you could offset the energy that is no longer expended by people travelling to meetings that are now held virtually

So why is HS2 needed?

From my view point I rarely physically travel north of Reading so HS2 isn't needed at all.

Others may disagree if their circumstances differ.

What we can be sure of is that if HS2 is under utilised then they will cut the time table to suit.

Article from that excellent website

http://www.lowtechmagazine.com/2015/10/ ... nergy.html

The internet is Jevon's paradox on acid. As the energy efficiency of data storage and transmission declines exponentially, the total energy consumed grows exponentially. So far, there is no physical constraint against this growth on the horizon other than the energy cost of streaming and transmitting ever higher resolution cute kitten videos.

Unfortunately, the latest technologies ( 4G and soon 5G) have much higher energy costs than wired or fibre transmission, and these are the fastest growing sectors of the internet.

(In fact, IIRC, it is already possible to power low energy devices by tapping the current electromagnetic fog that is the modern urban environment - to the ire of nearby smartphone users).

It is almost entirely the entertainment and propaganda functions of globalisation that consumes this energy, so with the collapse of globalisation we should expect this energy expenditure to fall, although I expect it to be as a result of the shut down of services rather than restriction to low energy uses like education, science and news reporting.

PS_RalphW wrote:Article from that excellent website

http://www.lowtechmagazine.com/2015/10/ ... nergy.html

The internet is Jevon's paradox on acid. As the energy efficiency of data storage and transmission declines exponentially, the total energy consumed grows exponentially. So far, there is no physical constraint against this growth on the horizon other than the energy cost of streaming and transmitting ever higher resolution cute kitten videos.

Unfortunately, the latest technologies ( 4G and soon 5G) have much higher energy costs than wired or fibre transmission, and these are the fastest growing sectors of the internet.

(In fact, IIRC, it is already possible to power low energy devices by tapping the current electromagnetic fog that is the modern urban environment - to the ire of nearby smartphone users).

It is almost entirely the entertainment and propaganda functions of globalisation that consumes this energy, so with the collapse of globalisation we should expect this energy expenditure to fall, although I expect it to be as a result of the shut down of services rather than restriction to low energy uses like education, science and news reporting.

Great site - thanks for the link. Interesting to see that 3G uses 15 times energy than WiFi and 4G more than 23 times as much.

I'm finding that more places are getting WiFi out in the open, especially in London where you can use a dedicated provider called Relish that uses WiFi supplied via lampposts. I met the owner of Relish once, and he come up with the idea as he didn't have the onerous demands from OfCom that the mobile providers get stuck with, so thought he'd exploit the rules

Relish operates its own 4G network. I hadn't heard that it had wifi access points as well.

It's also not doing very well unfortunately:

http://arstechnica.co.uk/business/2015/ ... lion-loss/

I find the invocation of Jevons a little bit puzzling (but then I find most invocations of Jevons puzzling).

The reason for the explosion in energy consumption is down to the increase in the number of connected people (we're adding something like 1 billion new internet users every 5 years), increases in data transfer rates, improvements in capabilities of connected devices, the provision of HD video streaming etc etc.

For sure if, as the article does, you compare vintage computers that consume hundreds of kW to modern mobile phones that consume hundreds of mW then you can say that we wouldn't have the number of devices now if we'd stayed with the old technology, but that's self evident.

Energy consumption is increasing in spite of not because of efficiency gains. The increase in energy consumption (as well as environmental concerns) is driving improvements in server centre energy efficiency, not the other way round.

That is not the only aspect. People now use smartphones, and increasingly 3G and 4G data services, as a background default activity to do whenever they are not doing anything else more entertaining like talking to the person next to them, waiting to go to sleep, waiting for the bus, etc, etc., and this forms an entirely new energy consuming activity which simply didn't happen before. If the hand held product wasn't cheap and easy to use, this energy consumption would not occur.

What I do is fold bits of paper at these times. I am excellent at origami, and there is almost always a waste piece of paper to hand, and the end product is still reusable or recyclable

Google's building windfarms.
http://googlegreenblog.blogspot.co.uk/2 ... -wind.html

PS_RalphW wrote:That is not the only aspect. People now use smartphones, and increasingly 3G and 4G data services, as a background default activity to do whenever they are not doing anything else more entertaining like talking to the person next to them, waiting to go to sleep, waiting for the bus, etc, etc., and this forms an entirely new energy consuming activity which simply didn't happen before. If the hand held product wasn't cheap and easy to use, this energy consumption would not occur.

What I do is fold bits of paper at these times. I am excellent at origami, and there is almost always a waste piece of paper to hand, and the end product is still reusable or recyclable

Yes this is very true but that's not due to the energy efficiency of the devices / internet itself. I have a bug bear about the Jevons and rebound and their widespread use as a stick to bash energy efficiency when it's almost always other factors that are leading to the increase in energy consumption (although i should note that rebound is very real and should be considered when dealing with energy efficiency).