2. What is FACEBOOK?
Facebook is a social networking site that makes
it easy for you to connect and share with your
family and friends online.
3. DID YOU KNOW?
Facebook handles upto 500 petabyte of data per day.
10 billion messages/day.
600 million photos/day.
4. Data Execute response
Query response times: 4ms reads, 5ms writes.
Rows read per second: 450M peak
Network bytes per second: 38GB peak
Queries per second: 13M peak
Rows changed per second: 3.5M peak
5. Which DBMS does Facebook use for saving
posts, comments and likes?
Facebook uses MySQL for the majority of the data
you would consider "Facebook data" like posts,
comments etc.
7. So, what are the other DBMS Facebook uses?
Hbase : for Graph search and messages
Cassandra : for inbox search
Haystack : for storing photos
Memcached : for an in-memory key/value store
Scribe : for logging
8. Growing Data Management
Hadoop: An open source Implementation of Google's
MapReduce
Hive: A system for querying datasets stored in Hadoop files
Cassandra: which provides a BigTable-like system for storing
massive amounts of structured data.
10. Small Data Challenges
Facebook’s News Feed requires 1000s of objects to render
and it is personalized for everyone on the service
TAO and Memcache are the primary systems that handle
caching our small data workload.
11. Big Data Challenges
ODS stores 2 billion time series of counters
Scuba is Facebook’s fast slice-and-dice data store
Hive is Facebook’s data warehouse with 300 PB data in 8 lakhs
tables.
12. What type of server does Facebook use?
For PHP -> HipHop for PHP.
Linux operating system -> Apache HTTP Servers.