Hadoop Related Jargons

Hadoop implements a computational paradigm named MapReduce. If you have no idea what is MapReduce, read the Wikipedia article here.

Some Hadoop Related Terms:

• Hadoop: developed by Y!, a map-reduce implementation
• HDFS: Distributed file system written in Java for the Hadoop framework
• Pig: High level scripting language to work with Hadoop
• Hive: Data warehouse infrastructure to work with Hadoop, uses HiveQL (an SQL-like language) 
• HBase: A non-relational key/value datastore to work with Hadoop
• Mahout: A set of machine learning algorithms to work with Hadoop on Big data

Some Other Systems Similar to Hadoop:

• Dryad
• Developed at Microsoft
• Tasks modeled as directed acyclic graph
• Sequential programs are connected using one-way channels
• S4
• Developed by Y!
• Stream processing
• Using Java Platform
• Spark
• Developed at UC Berekley
• In-memory queries, not just IO requests
• Implemented in Scala
• Needs a cluster manager (called Mesos)
• Storm
• Developed by Twitter
• Stream processing
• Guarantees message processing
• BashReduce
• works with Linux commands such as sort, grep
• Disco
• Developed at Nokia 
• Backend is written in Erlang
• Works with Pyton 
• developed at Nokia
• GraphLab
• Developed at CMU
• For machine learning tasks
• Data should fir in main memory
• Is not fault tolerant
• HPCC
• Uses Enterprise Control Language (ECL)
• In C++