theCUBE Live At EMC World 2014

EMC Corp. and Pivotal Labs reckon that data lakes are becoming more popular as scalable repositories for Big Data analytics, and have released a new scale-out Data Lake Hadoop Bundle 2.0 to address that need.

Data lakes work much the same way as natural lakes. Large amounts of information from multiple sources are stored there for later use. Like their real-world counterparts, data lakes must constantly be enriched with fresh information so that the contents don’t get stale.

Data lakes have two main benefits: they eliminate data silos and they can address the problem of how to analyze data stored in different formats. But data lakes need an underlying mechanism to maintain them, and without good metadata they can become what Gartner calls a “data swamp,” with so much bad data that quality control is lost.

EMC says its new Data Lakes bundle can accelerate customers’ predictive Big Data initiatives by providing an optimized configuration of scale out compute, storage and analytics software. The bundle “offers a high performance system for leveraging Hadoop for Big Data analytics,” said Jeremy Burton, President of Products and Marketing at EMC.

The release follows the first incarnation of EMC and Pivotal’s Data Lake Hadoop Bundle earlier this year, which married Hadoop and predictive analytics with scale-out storage.

The new bundle includes EMC’s high-performance Data Computing Appliance that’s designed to simplify the deployment and scaling of Hadoop. Other components include “a pre-tested, high performance Big Data analytic system featuring EMC storage”, and advanced analytics on Hadoop through enterprise appliances, adding up to a single, easy to implement solution, says EMC.

“Combining storage, compute and analytics capabilities for scale-out Data Lakes is tremendously valuable and the addition of predictive functionality means that customers can quickly put this solution to good use,” said Ashish Nadkarni, Research Director at International Data Corp.