Facebook has 0.5% of the installed base of Servers in World, hosted in about 10 colocation sites mainly in Santa Clara and Ashburn, VA. Google and Microsoft have more servers in more locations, but Greenpeace and other environmentalists don't find these companies nearly as interesting. Facebook represents its perception to media like OregonLive.
Q: Right after you announced your plans, Greenpeace and others chimed in about your power source (which includes PacifiCorp coal power). What did you learn about people's perceptions?
Jonathan Heiliger: In some respects it shouldn't be surprising, because we are a growing company and have become a fantastic target for people. That being said, we didn't explain well enough how efficient this site is relative to our current sites (Facebook currently leases data center space from other companies).
Facebook's new North Carolina Data Center is covered by the top technical news with many business journals and regular media picking up the news. Keep in mind what media is after is traffic. This isn't big news, and probably more news than Facebook wants as this type of news doesn't provide much business value to Facebook.
But this is enough news that Facebook's data center executives are in "damage control" of negative PR and Greenpeace is the top cause of the damage.
GreenFudge.org asks a good question.
Dirty coal and Green on Facebook. What’s the deal? And what should we do about it?
Recently Facebook launched their Green on Facebook page in an effort to green up their image after the big dirty coal data center debacle of earlier this year. According to their Facebook page, the Green on Facebook is
run by Facebook and will highlight our efforts to be a green and sustainable global citizen.
Image by Library of Congress (source: Flickr)
Together with 56.000 others, I became a fan op the page, and as many others I’m sure I’m pretty disappointed with the content of it. The wall is filled with links to various articles about different environmental topics, but very little information is available about Facebook’s own efforts to be sustainable and green. One article, supposedly from Facebook’s own engineers, discusses the topic of cooling strategies in data centers to increase energy efficiency. I could not stop but wonder if this post has anything to do with the new dirty coal scandal that Facebook is looking at today.
According to Jodie Van Horn, blogger for Greenpeace, Facebook has chosen a new data center location near Forest City, North Carolina that will – again – increase the demand for dirty energy. Greenpeace energy campaigner Gary Cook issued the following statement about this news:
Facebook is becoming a vehicle and poster child for change.
So yes what Facebook is doing is not OK. Greening up their image and at the same time opening a new coal fueled data center. But let’s not kid ourselves, every time we surf the web we are pushing CO2 into the air. And in this story as in any other, we are the customers.
So what should we do about it? Get off the Internet as long as it’s not sustainable? That’s one option but maybe not the smartest one. A better idea might be to make more conscious choices, on and off the Internet, and to become aware that no company, organization or corporation will ever change their ways before we do. So in the end, or for starters, even if we don’t stop using Facebook (which we should eventually do, but hey we’re all human), a good step might be to sign up for Greenpeace’s Unfriend Coal campaign, as long as we understand that signing a petition now and again is not our way to carbon free heaven. It’ll take much, much more than that to get there.
BTW, all this noise and issues with Carbon in Data Centers makes it much easier to discuss Green Data Centers.
Keep on telling the environmentalist your PUE, LEED building certification, energy efficient servers and your chiller-less cooling system.
People will learn these technical data center details are not what the public cares about. You can't change world talking about your Hadoop implementation.
It is not a secret anymore!
The Datawarehouse Hadoop cluster at Facebook has become the largest known Hadoop storage cluster in the world. Here are some of the details about this single HDFS cluster:
- 21 PB of storage in a single HDFS cluster
- 2000 machines
- 12 TB per machine (a few machines have 24 TB each)
- 1200 machines with 8 cores each + 800 machines with 16 cores each
- 32 GB of RAM per machine
- 15 map-reduce tasks per machine
That's a total of more than 21 PB of configured storage capacity! This is larger than the previously known Yahoo!'s cluster of 14 PB. Here are the cluster statistics from the HDFS cluster at Facebook: