Pivotal, Hortonworks Collaborate On Hadoop Management

Tech Toys For Summer Fun

Pivotal announced an alliance of sorts with Hortonworks on Monday, pledging to collaborate on development of the Apache Ambari project, open-source software used by Hortonworks to manage and monitor Hadoop clusters.

Pivotal said its pledge to collaborate with Hortonworks on Ambari is in keeping with its open source credentials and contributions to Cloud Foundry, Redis, Spring XD, and RabbitMQ, and with its "deep commitment" to Apache Hadoop.

"Pivotal is committed to supporting multiple approaches of installation technologies that will vary by use case, environment, and infrastructure decisions and requirements," said a Pivotal spokesperson in an email exchange with InformationWeek. "Ambari is one such use case. This is an example of Pivotal contributing to open source efforts to enable the entire Hadoop community and drive enterprise adoption."

[Want more on Pivotal's Hadoop plans? Read Pivotal Subscription Points To Real Value In Big Data.]

For now, Pivotal will continue to offer and develop Pivotal Command Center, which is the management software currently included with its Pivotal HD Hadoop distribution. Pivotal's spokesperson said the company will issue a detailed roadmap for Ambari and Pivotal Command Center in the third quarter.

Despite all the talk about open-source support, Pivotal also offers proprietary add-ons for Hadoop including HAWQ, a SQL-on-Hadoop option based on the Greenplum database, and GemFire XD, an in-memory SQL analysis on Hadoop option based on GemFire and SQL Fire. Both of these modules are managed by Pivotal Command Center.

"Ambari is extensible and we do plan to provide extensions to manage our proprietary offerings in the future," said the Pivotal spokesperson.

Pivotal's commitment is a boon to Hortonworks, which develops and promotes Ambari and includes it in its all-open-source Hortonworks Data Platform (HDP) Hadoop distribution. Hortonworks rivals Cloudera and MapR, by contrast, have proprietary management consoles that support certain commercial software components that are exclusive to their Hadoop distributions.

"We fundamentally believe the efforts of a vibrant open source community under the governance of the Apache Software Foundation (ASF) [are] the fastest path to innovation for Hadoop," said Shaun Connolly, Hortonworks' VP corporate strategy, in blog post.

When selecting servers to support analytics, consider data center capacity, storage, and computational intensity. Get the new Hadoop Hardware: One Size Doesn't Fit All issue of InformationWeek Tech Digest today (free registration required).