Apache Pig is a tool used in the Hadoop ecosystem that acts as an abstraction layer to MapReduce. Whereas Facebook created Hive to simplify the process of working with MapReduce and HDFS, Yahoo did the same thing with Pig.

• Apache Pig: Pig Latin is a high-level programming language similar to SQL
• Apache Hive: SQL-like Language called HQL 

MapReduce is the processing engine in Hadoop that breaks down big data into chunks so multiple machines can parallelize the processing. In doing so, the performance of storing and retrieving data in HDFS is drastically improved. The only problem is MapReduce requires Java programming skills, not the easiest language to work with.     

• Apache Pig: 1) Works with structured, semi-structured, and unstructured data 2) Used more by programmers
• Apache Hive: 1) Works with structured data 2) Used more by analysts

Apache Pig uses Pig Latin, a high-level textual language that abstracts away Java, making it easier for users to interact with MapReduce and HDFS using an SQL-like language. Supposedly, a dozen lines of code in Apache Pig could accomplish the same workload as a couple of hundreds of lines of code in MapReduce. 

Project Background

• Platform: Apache Pig
• Author: N/A
• Released: September 11, 2008
• License: Apache License 2.0
• Language: Pig Latin
• GitHub: apache/pig
• Runs on: Multi-platform
• GitHub Stars: 643
• GitHub Contributors: 11

Applications

• Perform various operations such as join, sort, filer, etc.
• Optimize execution of data automatically
• Create user-defined functions in languages
• Cycles a huge data volume
• Performs data handling in search stages

Summary

• It makes complex tasks easy to write, understand, and maintain with data sequence flow.
• It gives encoded permits to optimize the data execution automatically.
• Helps users to focus on semantics rather than efficiency.
• You can create your own functions for special-purpose processing.
• Twitter, LinkedIn, eBay, and Yahoo utilize Pig to deal with their big volumes of data.