Surprise! Digital Downloads Have Bigger Carbon Footprint Than Most Discs

blackrave

New member
Mar 7, 2012
2,020
0
0
ObserverStatus said:
blackrave said:
Only actual danger to environment is locally concentrated emission of large amount of carbon, due to acidic rains and similar short term effects
It's not just the rain that's becoming acidic, it's the whole damn ocean. I don't know if you've been reading the news lately, but entire marine ecosystems are collapsing.
#5

But it seems that most of it comes from shit we pump and dump into oceans, not atmosphere
 

CrystalShadow

don't upset the insane catgirl
Apr 11, 2009
3,829
0
0
direkiller said:
ilupir said:
The_Great_Galendo said:
Someone please explain how these numbers make any sort of sense.
Okay.

When you're downloading anything, you have to consider just how much traffic you're generating. The data has to go through several relays to get from the server to you. Now, the data itself won't be very costly, however, the demand that you (and your fellow gamers) are creating by downloading the games requires the internet relays to become bigger and bigger (at least in computing power).

As a sort of comparison, a communications engineer told me last year, that simply typing a single word into Google, and the resulting, will cost enought energy to bring a cup of water to a boil. (As this was something he told me, I haven't got a source, sorry)
Me thinks he was talking crap, not a little crap mind you, like supper mega crap.

They actually give the upper and lower bounds of energy use transmitting data.
it's .5-1.5 KWH/GB
KWH is equal to 3,600,000 joules and it takes 334,400 joules to bring a cup of water to boil from freezing.
So unless typing a word into Google takes up more room then most term papers he is wrong.
While I have no idea if it is correct or not, I wouldn't be so quick to assume it's wrong.

Google has a very large number of servers dedicated to it's search engine. And while it's improbable that any individual search request puts much of a load on the servers (after all, think about how many people must by trying to perform a search at any given moment), to be able to provide relevant results that aren't massively out of date, google's servers have to crawl and index as much of the web as they can as quickly as possible.
Estimates suggest what google actually manages to index may be as little as 10% of the whole internet, but even so this is hundreds of millions of web pages they have to check, run through indexing algorithms, and then somehow shove into a searchable database.
And they have to do this over and over, as often as they can afford to. (Otherwise the search results would be out of date and probably wrong.)

So if you take into account what has to happen behind the scenes to even make it possible to run a search in the first place, the aggregate cost of all the power used doing this could be quite substantial, even if the search request itself is not.

This cost doesn't even factor in data transmission, it's simply the cost of having thousands or even tens of thousands of servers running. (even if the average power use for one server is as little as 100 watts, which is optimistic considering typical PC power usage figures even at idle are often higher than this..., 10,000 servers would require 1 megawatt of power to run. It doesn't matter if this comes from renewable energy sources, the power is still needed to be able to run the servers.).
The usage per request then becomes something akin to the number of servers needed to run search functions + those doing crawling and indexing divided by the number of searches performed in a given period... (and energy usage per server)

If there are a million requests an hour, and 10,000 servers using 100 watts each, then the usage per request would be 1 watt hour per request. (3.6 joules).

But why guess? According to: http://www.internetlivestats.com/google-search-statistics/ - Google averages 40,000 searches per second.
Google won't say how many servers it has, but according to this page (http://www.datacenterknowledge.com/archives/2009/05/14/whos-got-the-most-web-servers/), it has at least 900,000. (based on energy usage estimates)

But since we have power usage figures (for instance here: http://en.wikipedia.org/wiki/Google_platform) of 500 and 681 megawatts. for their servers collectively, we can make some real estimates without knowing how many computers that actually is. (Of course, since it owns things other than search engines, it'd likely be lower than that.)

Making the slightly questionable assumption that this is only for their search engine, and not their web services (or youtube), we see that this works out to 500 megajoules for 40,000 searches (1 watt = 1 joule per second, and there are 40,000 searches a second). That's 12,500 joules per search. (Assuming none of the server capacity is for things other than searches.)

Now, what about boiling water? Well, 1 calorie is by definition as the energy needed to raise the temperature of 1 cubic centimetre of water by 1 degree Celsius. (1 calorie converts to 4.184 joules)
A typical cup is 250 ml, and 1 ml of water has a volume of exactly 1 cubic centimetre.

That means to raise the temperature of a cup of water by 1 degree would take 1046 joules. However, we need to do more than that... Assuming the water starts out frozen is a little dubious. That would be ice. unheated Tap water is anything from aboout 4 degrees to 15 degrees, depending on where you live, and the weather, but 12 is a typical average.

That means to boil it would require raising the temperature by 88 degrees. which is 92,048 joules.
(That's considerably less than your figures, but that's not so important given the other figures.)

Since 12,500 joules per search is probably towards the high end, and we can see that it is indeed not enough to boil a cup of water, we can definitely say it's probably wrong.
Then again, it may be an old fact from a period where it took more energy.
Alternatively, it would imply that for 4/5th of the day, no searches are taking place, but that's highly unlikely.

Anyway... Maths... What it does show though is that the two figures are at the same order of magnitude as one another. Which means the exact details of what takes place when you do a search could easily sway the result...