MapR had two major updates today: the company released the source code of its Hadoop distribution, and provided details on a fresh partnership with Canonical to develop a version of Ubuntu that will be pre-integrated with MapR M3, the community edition of its platform.
The source code of MapR Distribution for Apache Hadoop is available on Github, and includes enhanced versions of 12 Hadoop-related projects: Hive, Pig, Cascading, HCatalog, HBase, Oozie, Flume, Sqoop, Mahout and Whirr. The associated documentation and binaries can be downloaded from this freshly uploaded Maven repository.
In a second release that it published this morning, MapR revealed that its engineers are collaborating with Linux distributor Canonical to bundle M3 distribution with Ubuntu. The company is also working on Juju Charm that will allow OpenStack users to easily deploy MapR software in their environments.
The open-source trend, and by extension the Big Data movement, are not good for business if you happen to be a traditional enterprise vendor like Oracle. Jeff Kelly elaborated on the transition that the database market is undergoing in this month’s Wikibon newsletter.
“The problem for Oracle, as I outlined in a recent research note, is that the company’s scale-up, proprietary hardware and software, appliance-led model is at odds with the prevailing trend towards scale-out platforms and non-relational databases that use commodity hardware and open source software supported by some level of proprietary components. This movement is epitomized by Hadoop, the open source Big Data framework, but there are a number of non-Hadoop databases – and start-ups commercializing them – that are putting pressure on Oracle’s database sales.”
While some vendors are following behind the curve, others are doing their best to stay ahead of the market. This week Rackspace made its second NoSQL acquisition in a month, a Redis startup called Exceptional Cloud Services.