Facebook, Amazon, and Google -- three of the most visited websites in America. To the millions of visitors who land on the home page of these sites, the technology and infrastructure that works behind the scenes is the probably the last thing that crosses their mind.
The details on specific hardware and software with companies like these can be hard to find. In his book Planet Google, Randall Stross, wrote that Google goes to “extraordinary lengths to keep the company's hardware hidden from view. The facilities are not open to tours, not even to members of the press.” There is however, just enough information available to get an idea of what it takes to keep millions of visitors happy and coming back for more.
And what makes websites like these work, well while being accessed by millions of visitors at once, is nothing short of incredible. The size and scope of these websites are lost on its visitors because, frankly, they work. According to data from Alexa.com, the average page load time for Facebook, Google, and Amazon is less than 2 seconds.
According to data from Google, Facebook serves roughly 260 billion page views per month and is estimated to use 60,000 servers. In a 2009 interview, Jeff Rothschild, the Vice President of Technology at Facebook, discussed the company’s infrastructure, in which he said that Facebook manages more than 25 terabytes of data per day just in logging information. He also shared some large numbers associated with Facebook’s photo storage operation, which now stores over 80 billion images (20 billion images, each in four sizes). Rothschild went on to say that real challenge isn’t storage, but delivery. ”We serve up 600,000 photos a second,” he said.
In June of 2010, Facebook announced plans to complete the initial phases of its first company-owned data center in Prineville, Oregon. In July, they announced plans to double the size adding another 160,000 square foot to the original 147,000 facility.
Amazon grew from a tiny online bookstore into one of the largest stores in the world. Looking for something? Chances are you’ll find it on Amazon. It’s estimated that Amazon has more than 55 million active customer accounts and more than 1 million retail partners worldwide. To display a page on Amazon’s website, it uses between 100 – 150 services.
Amazon’s data centers are located all over the world and handle more than 72 items ordered every second. Amazon has built their own cloud infrastructure and offers their data services to anyone through Amazon Web Services. They also are behind many of the world’s well-known online stores like Target, Toys-R-Us, Land’s End, and Nordstrom.
When you connect to Google, you are connected to one of the world’s most complex content delivery networks. Google’s server infrastructure includes load balancers that take every request and forward it to one of the web servers running Google’s proprietary web server software through “Squid” proxies, which enable repeated requests to be sped up by caching user information.
Google has numerous data centers around the world with at least 12 significant installations in the United States. The largest of these located in Oregon and Codenamed “Project 02”, is approximately the size of two football fields with cooling towers that are four stories high.
It’s always fascinating to me when someone makes something hard look easy. Whether it’s Michael Jordan or Jeff Bezos, it still has the same effect. And it still makes me smile.
Director of Magic
RT @thundertech: Welcome to #CLE for #KenticoPartnerSummit... @Kentico, @TeamBizStream, @iMediainc, @silvertech, @BitwiseIndiana, @ucommerc…
@OCSOALERTS @FLGovScott @Rep_Mel_Ponder Great to have you all join us!