HadoopDB
An Architectural Hybrid of MapReduce and DBMS Technologies for Analytical Workloads.

HadoopDB is:

1. A hybrid of DBMS and MapReduce technologies that targets analytical workloads
2. Designed to run on a shared-nothing cluster of commodity machines, or in the cloud
3. An attempt to fill the gap in the market for a free and open source parallel DBMS
4. Much more scalable than currently available parallel database systems and DBMS/MapReduce hybrid systems.
5. As scalable as Hadoop, while achieving superior performance on structured data analysis workloads

For more detail, check out the DBMS Musings blog post, or the paper below.

The Paper:

HadoopDB: An Architectural Hybrid of MapReduce and DBMS Technologies for Analytical Workloads. Azza Abouzeid, Kamil Bajda-Pawlikowski, Daniel J. Abadi, Avi Silberschatz, Alex Rasin. In Proceedings of VLDB, 2009. [PDF]

News:

• Here is the latest paper on performance enhancements in HadoopDB published at SIGMOD'2011.
• HadoopDB technology was commercialized into Hadapt.
• Daniel Abadi was invited to give a talk at Hadoop World: NYC 2010.
• Our demonstration of HadoopDB in Action was accepted for SIGMOD 2010.
• We are invited to present HadoopDB at New York Hadoop User group meetup on April 21.
• HadoopDB was presented at New England Database Summit 2010 during the poster session.
• Call for contributions is posted here.
• HadoopDB's presentation at CLuE PI Meeting 2009 in California is now available.
• The Hadoop World: NYC 2009 presentation is now available.
• Our VLDB 2009 slides are here.

Press:

• Computerworld
• O'Reilly Radar
• Dr.Dobb's
• StorageMojo

Links:

Get HadoopDB Now!

• Quick Start Guide [HTML]
• Java Documentation [HTML]

Project Members

• Azza Abouzeid
• Kamil Bajda-Pawlikowski
• Daniel J. Abadi
• Avi Silberschatz