Pig is a data processing tool that is part of the Hadoop suite. It provides for the writing of scripts executed on the Hadoop infrastructure without having to first write Java tasks using the MapReduce framework. In addition, it has functionalities for loading data from an outside source to the HDFS cluster and others for exporting data for use by third party applications.
Pig is based on its own language, Pig Latin. It also provides access to Java's application layer. Pig Latin is a fairly simple language that allows developers from backgrounds other than Java to produce processing scripts that can run on Hadoop far more quickly.
In practice, Pig is chiefly used to load external data to HDFS files and to convert files in order to facilitate their analysis, especially in cases involving multiple steps (due to the procedural nature of the language and its ability to store temporary results).
This solution does not have any opinions yet