Skip to main content

Hadoop Graduates to Version 1.0

Recently, version 1.0 of MapReduce framework and distributed computing platform was released by the Apache Hadoop developers.

Hadoop's success can be attributed to the platform's capability to manage enormous data sets and handle complex data processing. Hadoop is also capable of distributing tasks between thousands of nodes (maps) and assimilating results from those nodes (reduce).

Hadoop version 1.0 has been built on security branch of Hadoop 0.20 and is expected to be more stable and reliable. Scalability is definitely a plus point here as 50,000 nodes installations have already been deployed successfully.

Hadoop's origin can be traced back to Yahoo! where it was created by Doug Cutting to support Nutch search engine. Research related to Google's MapReduce and Google File System served as an inspiration for the platform.

Later on, the system was contributed to Apache Software Foundation. There it became a major project for the organization's ecosystem. The project took care of many sub projects which were addressing many different aspects that are part of the challenge faced while distribution of computing.

Slowly and gradually many companies were set up around Hadoop like oCouldEra and Hortonworks and in fact companies like eBay, Twitter, StumbleUpon, Facebook, IBM, LinkedIn and Adobe also started using and contributing to the Hadoop project, reported The H Open.