Just 75 days after buying Greenplum, EMC on Wednesday announced an appliance based on the acquired vendor's massively parallel processing database. The new appliance is said to offer lower prices and faster data-loading speeds than its competitors. It also offers unique backup and recovery options.
The EMC Greenplum Data Computing Appliance combines the Greenplum 4.0 database, Dell servers, Brocade switches and direct-attached disk storage. Half-rack and full-rack appliances will be assembled, preinstalled with software, and shipped from EMC's factories in the U.S. and Cork, Ireland.
Prices will start at $1 million for a half-rack appliance with 18-terabytes of user-available storage. By applying compression, EMC said customers can store up to 72 terabytes. That brings prices below the less-than-$20,000-per-terabyte level claimed by Netezza last year. EMC-rival IBM is currently in the process of acquiring Netezza.
This is not the first time Greenplum's database has been offered as part of a data warehousing appliance. It was previously packaged with Sun server and storage technologies, but Sun's early 2010 acquisition by Oracle put the kibosh on that deal. Oracle competes with Greenplum (and now EMC) with its Exadata appliances.
EMC's new appliance excels at data-loading, with a rating of 10 terabytes per hour. That's twice as fast as Oracle Exadata systems and five times faster than Teradata and Netezza offerings, according to EMC. For most purposes, other vendors' load speeds are fast enough, according to independent analyst Curt Monash, but some network monitoring and logging use cases demand the highest load speeds available, he said.
The appliance offers two EMC-contributed storage options the likes of which rivals don't match. The first option is an integration with EMC DataDomain, a secondary storage and backup and recovery appliance that can deduplicate data, backup at designated times, provide point-in-time copies and reduce recovery times. The second feature is built-in integration with EMC RecoveryPoint, which gives users of that software the option of replicating appliance data onto EMC (or third-party) storage area networks for disaster recovery purposes.
The EMC Greenplum Data Computing Appliance is the first offering from a newly established Data Computing Products Division of EMC. EMC will continue to offer the Greenplum database separately, with certified configurations on Dell and HP hardware.
The company said that in the future it might look beyond the data warehousing arena (a realm Bob Evans explores in this article about the prospects for optimized systems moving up the stack into applications). But for now, EMC is concentrating on data warehousing, as are Teradata, SAP/Sybase, SAS, Vertica and few other vendors.
Oracle already has appliances in the transactional arena, and IBM, HP, SAP and Microsoft have signaled near-term interest in broader "optimized systems" offerings.