Hadoop Hive Gets Production Polish
Edging ever closer to enterprise software maturity, the Hive data warehouse software has been updated with new capabilities considered essential for production use, such as indexing, concurrency and advances in authentication management.
"We're pleased that the Hive release includes advances in performance, security and usability that our enterprise customers and business intelligence partners can benefit from," said Charles Zedlewski, a vice president at Hadoop platform provider Cloudera, in an e-mail interview.
A component of the Apache Hadoop data processing framework, Apache Hive is open source software for running data warehouse-styled operations against large datasets stored in Hadoop file systems. It provides such operations as data summarization, ad-hoc querying, and analysis. Volunteer developers from Cloudera, Facebook and other companies contribute to the code-base.
Interest in Hadoop has been growing of late, as many see the platform as an essential tool for analyzing large amounts of data.
Version 0.7.0 of the software, released Tuesday, is the first major upgrade since version 0.6 was released in October.
The freshly released version contains a multitude of new features, many of which have long been available in commercial databases and data warehouses.
One new feature is indexing. With indexing, a summary of the data set is created that the system can use to speed processing of lookup and range queries. Authentication has been strengthened as well, allowing Hive to integrate with authorization information from other repositories.
Hive 0.7 also features a new concurrency model. Prior versions of the software had no safeguards in place to prevent the system from updating a piece of data at the same time that data is being read by another part of the system. As a result, the system may return incorrect data. The new concurrent model prevents access to a piece of data that is being updated at that moment.
Hive version 0.7.0 works with Hadoop 0.20.1 and 0.20.2.