Extensible-record stores (also called column stores) were initially motivated by Google’s Big Table project. In the system, data is considered in tables with rows and column families, in which both rows and columns can be split over multiple nodes:
- BigTable: Big Table was introduced by Google in 2004, as a column store to support various Google services. Big Table is built on the Google File System (GFS) and can easily be scaled up to hundreds and thousands of nodes, maintaining Terabytes and Petabytes scale of data.
- HBase: HBase is an Apache open source project and was developed in Java, based on the principles of Google’s BigTable. HBase is built on the Apache Hadoop framework and Apache Zookeeper, in order to provide a column-store database. As HBase was inherited from BigTable, they share a lot of features, in both their data models and architectures.
- Cassandra: Cassandra is an open source NoSQL database that was initially developed by Facebook, in Java. It combines the ideas of both BigTable and Dynamo, and it is now open sourced under the Apache license. Casandra shares the majority of its features with other extensible record stores (column stores), in both data modeling and functionality.