The folks at Hortonworks have been hard at work. Beyond the company's leadership role in YARN, the component that frees Hadoop from the tyranny of batch processing, Hortonworks has been managing a huge effort to retrofit Hive to work with it, in order to bring Hadoop more into the Enterprise mainstream.
Big improvements for Hive
With the 2.1 release of Hortonworks Data Platform (HDP) being announced today, the company has completed that Hive interactive query capability — known as Stinger Phase 3 — and is simultaneously releasing it for both the Windows and Linux platforms. This is a big advance. We knew it was coming, and now it's here.
What we did not know was that HDP would include a slew of other new components. These include Solr for search-based operations; Storm, for steaming data processing; a preview of integrated Spark for distributed in-memory processing; the Knox security component, which provides cluster-level authentication and authorization services as well as GRANT and REVOKE functionality in Hive; and Apache Falcon for data set replication, retention, audit and lineage.
The inclusion of Solr into HDP goes beyond mere integration of another open source component. In fact, Hortonworks is announcing a new reseller partnership with LucidWorks, arguably the most important commercial entity behind Solr and Lucene.
HDP 2.1 is available as a preview today and will be released to general availability on April 22nd.
This week is turning out to be a real watershed in the Big Data world, with the feature-rich Hortonworks Data Platform 2.1 reveal; the announcement of the Pivotal Big Data Suite; and Cloudera's announcement of a $900M round of capital, with Intel taking an 18 percent stake in the company. We're only one quarter into the year, and we should expect lots more action before the calendar rolls over to 2015.