May 09, 2011, 4:00 PM — LAS VEGAS - EMC Monday unveiled a purpose-built appliance for processing both structured and unstructured data sets for business analytics tasks.
EMC today also announced the availability of two new business intelligence software products -- the Hadoop-based EMC Greenplum HD Community and Enterprise Editions -- at its EMC World user conference here.
Service contracts for both software products includes installation, training and global technical support.
The Greenplum HD Community Edition is a fully-certified downloadable free software stack. The software is based on Hadoop, an Apache data management software, and is optimized to run on virtual machines.
Greenplum HD Enterprise Edition is tailored for corporate data centers, with capabilities like fault tolerance through automated node failure detection and notification, multi-site management and data management features such as snapshots and wide area replication.
It also offers simple data loading from databases and access via a native Network File System (NFS) interface.
EMC claims its version of Hadoop delivers two to five times the performance over the standard packaged versions of Apache Hadoop.
EMC had signed an agreement with Cloudera last fall to use its Hadoop-based data management software and services.
However, Scott Yara, vice president of products with EMC's Data Computing Division and co-founder of Greenplum, said today that the company is moving in a "new direction" with partner MapR Technologies, a start-up in development mode for the past two years.
MapR built a proprietary replacement for the Hadoop Distributed File System (HDFS) that can replace existing installations of the Hadoop file system.
John Schroeder, CEO of MapR, said his company's version of map reduce technology returns far faster data analytics results, and can manage larger data sets on fewer machines than current Hadoop iterations.
"We can reduce the size of a cluster," he told a gathering of reporters and analysts gathered at the show. "That's a tremendous TCO savings."
Luke Lonergan, CTO of EMC's Data Computing Division and another co-founder of Greenplum, added that EMC is working with dozens of resellers to get the MapR Hadoop software to customers. The distribution channels should go live later the second quarter of this year.
No pricing has yet been released for the Enterprise-edition.