At the Web 2.0 Summit in San Francisco, Mike Schroepfer, Facebook’s VP of Engineering talked a bit about the infrastructure of the company. Schroepfer’s mentioned that Facebook is not afraid to be bold and take risks to get done what it needs to from an engineering perspective. They’ve come up with a number of their own solutions to solve a massive scaling problem, and sometimes we (the users) see bugs as a result of that. But, Facebook’s relative stability is pretty impressive. Schroepfer threw out some huge numbers. Among them: * Users spend 8 billion minutes online everyday using Facebook
* There are some 2 billion pieces of content shared every week on the service
* Users upload 2 billion photos each month
* There are over 20 billion photos now on Facebook
* During peak times, Facebook serves 1.2 million photos a second
* Yesterday alone, Facebook served 5 billion API calls
* There are 1.2 million users for every engineer at Facebook
* Facebook connect is growing even faster than Facebook
* In over a year we went from 100 million daily actives to 300 million daily actives
* When Zuckerberg started the site, it was only Harvard, then more schools were added, but they were silos. Then we connected those schools, then in 2006 we opened it up.
* It’s way bigger than a single database, so you have to spread this data out.
* Rendering a homepage on Facebook, which do it a couple billion times a day, in a couple of seconds.
* We use something we built which we call “multifeed” which we use along with memcache to get data in milliseconds.
* 50 million operations a second via memcache
* We scaled memcache 5 X its original performance – we rewrote it
* Importance of our culture
* Move Fast – sometimes we push bugs, but that’s innovation Be Bold – change something if we need to. Source: Tech crunch
No comments:
Post a Comment