Secret Facebook's system architecture

Web front end is written in PHP. Facebook's HipHop [1] ambition turn into a C + + PHP and use g + + compiler, so that templates can commemorate the business logic wafer and the Web to provide lofty extravaganza.
Service business logic in the form, the use of Thrift [2]. The Service according apt the assorted absences of the PHP, C + + alternatively Java implementation (you can likewise use a digit of additional languages, ...) written in Java and Services does no use anyone enterprise application server, merely use a Facebook own custom applications server. Looks like it re-invent the cycle,Equilibrio Xtreme, merely these are only exposed apt the Thrift Services use (the digit namely so extensive),PULSERA TRANSPARENTE CON LOGO XB, Tomcat is also heavy, and may even be too Jetty point, the joined value needed apt Facebook does no make sense. Persistence along the MySQL, Memcached [3], Facebook's Cassandra [4], Hadoop's HBase [5] completed. Memcached uses a MySQL memories Cache. Cassandra Facebook engineers agree that their use is decreasing, for they favor HBase, for it's simpler consistency model to its MapReduce skill. Offline processing using Hadoop and Hive. Log, press, feeds file to use Scribe [6], its aggregation and there HDFS, its use Scribe-HDFS [7], allowing the use of MapReduce for stretched analysis. BigPipe [8] is their custom technology used to build up sheet display. Varnish Cache [9] as a HTTP agent. They use this for of speed and efficiency. [10]. Gaoding users to upload 1 billion photos stored, its handling at the Haystack, Facebook has amplified its own Ad-Hoc store a program, its effectively to do some low-level optimization and the ]. Facebook Messages with their own building, its apparently built ashore a dynamic bunch infrastructure. Business logic and persistence are encapsulated in a so-called 'Cell'. Each 'Cell' are handled as portion of the user, the new 'Cell' can be additional as way Redu [12]. Persistent archive using HBase [13]. Facebook Messages are stored in the search engines by the HBase an inverted concordance construction. [14] Facebook search engine implementation details as distant as I understand the current status is unknown. Typeahead search using a custom cache and retrieval logic. [15] Chat based aboard a Epoll server, the server amplified in Erlang, the Thrift access [16]
afford to the components on those resources,PULSERA NEGRA CON VENEZUELA, the emulating is some message and quantity, but there are some unknown :
Facebook is estimated that extra than 60,000 servers [16]. Their latest data hub in Oregon Prineville,PULSERA BLANCA CON NETHERLANDS, its completely custom design based on hardware [17] it was only recently publicly Open Compute project [18]. 300 TB of data exists to handle with Memcached [19] their Hadoop and Hive clusters formed by the 3000 servers, every server has 8 cores, 32GB of memories, 12TB of disk, all have 24,000-CPU-core, 96TB 36PB of memory and hard drive. [20] 100 billion a daytime volume of kicks, 500 million photos, 100 billion kicks per day,PULSERA BLANCA CON LOGO XB, 50 billion photos, 3 万亿 objects are Cache,XB 2011, annual 130TB of log (July 2010 data) [21 ]
references
[1] HipHop for PHP: http://developers.facebook.com/blog/post/358
[2] Thrift: http:/ / thrift.apache.org /
[3] Memcached: http://memcached.org/
[4] Cassandra: http://cassandra.apache.org/
[5] HBase: http://hbase.apache.org/
[6] Scribe: https://github.com/facebook/scribe
[7] Scribe-HDFS: http://hadoopblog.blogspot.com/2009/06/hdfs-scribe-integration.html
[8] BigPipe: http://www.facebook.com / notes/facebook-engineering/bigpipe-pipelining-web-pages-for-high-performance/389414033919
[9] Varnish Cache: http://www.varnish-cache.org/
[10] Facebook goes for Varnish: http://www.varnish-software.com/customers/facebook
[11] Needle in a haystack: effective storage of billions of photos: http : / / www.facebook.com/note.php?note_id=76191543919
[12] Scaling the Messages Application Back End: http://www.facebook.com/note.php?note_id=10150148835363920
[13] The Underlying Technology of Messages: https://www.facebook.com/note.php?note_id=454991608919
[14] The Underlying Technology of Messages Tech Talk: ; http://www.facebook.com/video/video.php?v=690851516105
[15] Facebook's typeahead quest architecture: http://www.facebook.com/video/video.php ? v = 432864835468
[16] Facebook Chat: http://www.facebook.com/note.php?note_id=14218138919
[17] Who has the most Web Servers?: http://www.datacenterknowledge.com/archives/2009/05/14/whos-got-the-most-web-servers/
[18] Building Efficient Data Centers with the Open Compute Project: http://www.facebook.com/note.php?note_id=10150144039563920
[19] Open Compute Project: http://opencompute.org/
[20] Facebook's architecture presentation at Devoxx 2010: http://www.devoxx.com
[21] Scaling Facebook to 500 millions users and beyond: http://www.facebook.com/note.php?note_id = 409 881 258 919