MapR announces version 1.2 of its distribution for Hadoop
MapR has announced version 1.2 of its distribution for Apache Hadoop for managing and analysing large quantities of data. The new version includes the foundations for supporting MapReduce 2.0 once it is ready for production use. It also incorporates a C/C++ libhdfs implementation for higher performance access to the Hadoop distributed file system.
Native support and installation packages for Windows and Mac OS X clients have also been introduced, allowing them to easily run Hadoop applications without needing to install third-party software such as the Cygwin libraries on Windows. The new MapR version upgrades several packages including HBase, the Hive SQL database, and the Pig programming language; the updated version of HBase, 0.90.4, has had its stability improved through the backporting of fixes by MapR. The developers have added a VMWare virtual machine to the release to allow users more flexibility to experiment with MapR's capabilities.
Version 1.2 of the MapR distribution will be available to download this week. MapR is available in two versions: the M3 edition is free and is supported through the community, the M5 edition is provided on a subscription basis, with full support, and has additional features including mirroring and snapshots. A thirty day trial version of M5 is available for those who want to evaluate the commercial features.