Facebook y su correlación con Big Data y Hive

Abstract:

We live in an era of “Big Data” produced by a wide range of applications, but facing the problem of scalability and high speed access. This paper focuses its analysis in this type of situation, and provides special support for batch oriented analytics applications. Facebook Data Infrastructure supports a wide range of applications, often this infrastructure is built on open-source technologies such as Apache Hadoop, Hadoop Distributed File System (HDFS), MapReduce and Hive. In this work the main idea is to understand how MapReduce facilitates parallel databases accessibility, include storage-system independence and fine-grain fault tolerance for large data sets. The main contribution is to show future works regarding this area.

Año de publicación:

Keywords:

Fuente:

google