There is a piece in today’s Crikey email about this story on the carbon impact of each individual Google search that has been doing the rounds of the papers and blogosphere today. In the email, we discuss the problem with the article — that the scientist behind the study is actually promoting his business and the journalists happliy play along — but really, there is a second issue with this article.
Even if the article wasn’t largely a veiled PR piece for the scientist’s company, there are still problems.
Performing two Google searches from a desktop computer can generate about the same amount of carbon dioxide as boiling a kettle for a cup of tea, according to new research.
While millions of people tap into Google without considering the environment, a typical search generates about 7g of CO2 Boiling a kettle generates about 15g. “Google operates huge data centres around the world that consume a great deal of power,” said Alex Wissner-Gross, a Harvard University physicist whose research on the environmental impact of computing is due out soon. “A Google search has a definite environmental impact.”
It’s true — a Google search, or any electronic function you perform, comes at a carbon cost. But a) Why single out Google searches above emails, Twittering, Instant Messaging or that app that turns your iPhone into a light saber, and b) Why put the focus on individual computing tasks at all, when surely the real issue is the infrastructure behind those functions and whether that is energy efficient?
TechCrunch have picked up on this issue, and offer some good analysis:
There’s no doubt that Google consumes a massive amount of energy, with hundreds of millions of searches conducted every day and data centers scattered across the globe. But let’s try to shed a little perspective on things.
A single book runs around 2,500 grams of CO2, or more than 350 times a Google search. By some estimates, a single cheeseburger has a carbon footprint of around 3,600 grams — over 500 times larger than a Google search. Granted, meat in general has a notoriously large carbon footprint, but if you’re genuinely concerned about your environmental impact then try cutting a burger from your diet every week and search guilt-free (you may even lose a few pounds).
Over at Slashdot, many of the geeks are questioning the data itself. Says one commentor:
I think their main folly is that they don’t distinguish between the power necessary to service requests vs. the total power used (which includes all the power it takes to index sites and store the results so they can be fetched quickly etc.) There is a big difference as the power required to index is relatively static and thus doesn’t depend on the number of searches. In fact, the power per search using their methodology may actually drop the more searches that are performed because each search’s share of the power required for indexing drops.
Whilst another notes:
Here’s some math:
250g water in a cup of tea.
Specific heat of water = 4186 J/kg/(degree C). (see http://en.wikipedia.org/wiki/Specific_heat_capacity [wikipedia.org]).
80 Celsius degree change from room temperature to boiling.
To boil a teacup’s worth of water, therefore it takes ~80 kJ.
For this to be twice the energy consumed with one search, that’s ~40 kJ per search.
If a search takes Google about 100 ms, that means Google would be using 400 kW while responding to your search. That feels like it’s about 3 orders of magnitude too high. It’s possible that the original researchers got Calories and kCal confused.
Certainly it IS important for people to recognise that each individual electronic task has its own carbon cost, and to that end, the gaggle of Google-searches-are-killing-the-planet articles make a salient point. But ultimately, it seems to be missing the forest for the trees.