EMC doubles performance of Data Domain dedupe arrays

Vendor also unveils a data deduplication archive array for long-term backups of data

EMC today unveiled upgrades to its high-end and midrange Data Domain deduplication arrays, boosting performance with new Intel processors, and capacity by using new drives.

The company also announced the Data Domain Archiver, which acts as an online storage repository for long-term backups, or data kept longer than typical 90 or so days.

The single-controller Data Domain Archiver array has a maximum data throughput rate of 9.8 terabytes per hour and can store a whopping 768TB of raw capacity, or 28.5 petabytes of deduplicated data, according to Rob Emsley, senior director of product marketing at EMC's Backup Recovery Systems Division.

The DD Archiver supports all current Data Domain tools, including the DD Replicator, a data replication application that copies files between Data Domain appliances, and the DD Retention Lock software , which saves files in a non-rewriteable and non-erasable format.

"We introduced a policy into the Data Domain Archiver ensuring that as backups age, say longer than 90 days, we move the data associated with it to other parts of the storage system or the archive tier," Emsley said. "We fill up those archive trays of storage and then seal them and have them remain on line for retrieval, not from tape but from disk."

EMC's refresh of the DD880 and DD690

EMC's refresh of its high-end DD880 array, now called the DD890, adds an Intel Nehalem processor, which allows the box to ingest data at up to 14.7TB per hour. Its predecessor has a data intake rate of up to 8.8TB per hour.

The DD890 now has a total raw capacity of 384TB, twice the DD880's 192TB capacity.

Using its deduplication algorithm, the appliance is able to theoretically store up to 14.2 petabytes of data, Emsley said.

EMC's new DD860, which supersedes the DD690, adds an Intel Westmere processor, which allows it to ingest up to 9.8TB of data per hour. The deduplication device now has up to 192.2TB of raw capacity. Its predecessor ingested up to 1.1TB of data per hour and had a data capacity of up to 48TB.

With deduplication, the DD860 could theoretically store up to 7.1 petabytes of data.

Emsley said the upgrades mean more users can protect more data faster.

"Clearly, if you're talking about backup storage, speed matters. With shrinking backup windows, the more terabytes per hour, the more the system can protect from a backup perspective," he said.

The new DD690 and DD880 arrays are single controller devices that allow trays of serial ATA (SATA) disk to be added for up to 7.1 petabytes and 14.2 petabytes of data, respectively.

The company also announced an update to its dual-controller configuration Global Deduplication Array (GDA) that it announced last April . The GDA is a dual configuration of the DD890 array, providing twice the throughput and capacity. The GDA is now capable of ingesting up to 26.3TB per hour and can store up to 570TB of capacity.

The updated GDA also adds support for virtual tape libraries (VTL), or disk arrays that pose as tape libraries to backup servers. The GDA acts as a target device to the VTL, which allows it to act as longer-term storage for backups.

The first iteration of the GDA could be used with Symantec NetBackup and EMC NetWorker backup software. "Now the Global Deduplication Array now supports IBM's Tivoli Storage Manager because it supports VTL," Emsley said.

Emsley said the GDA also supports IBM i products , or the unified operating system that run its open systems servers, formerly known as AS/400s.

"Now a customer considering using Data Domain for his open systems workloads can leverage the same system for an IBM i workload," he said.

A DD890 controller with no storage capacity has an entry list price of $300,000. The DD860 has a starting price of $160,000. The entry level DD Archiver, which consists of its controller and one storage shelf of disks in the active tier, lists for $222,000.

Lucas Mearian covers storage, disaster recovery and business continuity, financial services infrastructure and health care IT for Computerworld. Follow Lucas on Twitter at @lucasmearian , or subscribe to Lucas's RSS feed . His e-mail address is lmearian@computerworld.com .

Read more about storage in Computerworld's Storage Topic Center.

Copyright © 2011 IDG Communications, Inc.

7 inconvenient truths about the hybrid work trend
Shop Tech Products at Amazon