I love occasional glimpses behind the scenes. In The Wizard of Oz, I would have been Toto, pulling back the curtain to figure out what was running the big head. The New York Times gives us a fascinating look at the data centers powering the online services that drive our world – Google, Microsoft, Amazon, and more. Most of us have only the vaguest idea that somewhere there must be a lot of computers indexing the web and showing us Google search results. Not only are there a lot of computers, the scale of it probably exceeds anything you would guess.
A data center will have servers in racks. There are Google data centers that each have as many as 45,000 servers humming quietly, using electricity and generating heat. In fact, electricity and heat are some of the biggest considerations for the buildings chosen or built as data centers.
Data centers worldwide now consume more energy annually than Sweden. And the amount of energy required is growing, says Jonathan Koomey, a scientist at Lawrence Berkeley National Laboratory. From 2000 to 2005, the aggregate electricity use by data centers doubled. The cloud, he calculates, consumes 1 to 2 percent of the world’s electricity.
Much of this is due simply to growth in the number of servers and the Internet itself. A Google search is not without environmental consequence — 0.2 grams of CO2 per search, the company claims — but based on E.P.A. assumptions, an average car trip to the library consumes some 4,500 times the energy of a Google search while a page of newsprint uses some 350 times more energy. Data centers, however, are loaded with inefficiencies, including loss of power as it is distributed through the system. It has historically taken nearly as much wattage to cool the servers as it does to run them.
As our world grows in complexity and speeds up, one issue is the length of time it takes for a server to transmit data back and forth to you. If the server is closer, the information gets to you faster – the bits are limited to the speed of light. Financial companies will build their data centers near the stock exchanges to reduce the lag by a few milliseconds. According to the article, “it is estimated that a 100-millisecond delay reduces Amazon’s sales by 1 percent.”
Amazon has built data centers all over the world. A few years ago, the company realized it had set up such a robust and redundant system that it began to set up methods for outside developers to write programs that could use Amazon’s storage space and processing power. The result, Amazon Web Services, has become a huge resource for third parties, yet it’s virtually unknown to the outside world. One of the AWS divisions, Amazon Simple Storage Service (“Amazon S3”), can be used for online file storage and backups; in March, it was holding 52 billion objects.
One of my favorite statistics: at the beginning of 2008, Amazon Web Services was using more bandwidth than all of Amazon’s web sites put together, including all of their retailing operations.
It all puts the single server in your small business’ utility closet into perspective, doesn’t it?