Dell, Quantum, EMC line up on de-duplication

Dell will enter the data de-duplication business next year with products that are based on Quantum's software and compatible with EMC's gear.

The products, which will feature Quantum's software on dedicated hardware boxes, will start to come out in the first half of next year and eventually will span all of Dell's storage lines from small and medium-size business gear to large enterprises equipment. Along with Quantum and EMC, Dell will have a common software architecture that will make products from all three companies compatible, Dell said.

Data de-duplication is designed to slash the requirement for storage capacity by finding identical files or bits of information in an organization's stored data and saving them just once. For example, if two versions of a presentation differ only in the details of one slide, de-duplication lets an enterprise save just one copy each of all the slides that are identical.

Major storage vendors, including Hewlett-Packard and NetApp as well as EMC and Quantum, already sell de-duplication products. Dell believes its entry now will allow it to start out with more mature products, said Brett Roscoe, senior manager of Dell's storage division.

The alliance with Quantum and EMC is a natural move, given that EMC already uses Quantum software for data de-duplication and is also an OEM (original-equipment manufacturer) partner of Dell. But it creates a compatibility that may allow customers to implement de-duplication widely with less cost and trouble. Users will be able to replicate data to and from products from the three vendors, at multiple sites, while taking advantage of the storage efficiency of de-duplication, Roscoe said.

The technology will also help make disk storage competitive with tape. Organizations typically use tape for disaster recovery, which at a dollar per gigabyte is much less expensive than disks, but it's usually harder to get data back off tapes when needed, Roscoe said. With the compression that data de-duplication brings, the cost per byte comes down. Customers may use disk storage for recently created data, such as material from the past 90 days, that end-users most frequently need, he said.

Dell's gear will use "variable block" de-duplication, which can examine data within both files and smaller blocks of data and find duplicated bits, Roscoe said. It is the most processing-intensive form of de-duplication but can achieve the highest levels of compression, he said.

"You're slicing it finer and finer, and the finer you slice it, the more you're going to find in terms of redundancies," Roscoe said.

The IDG News Service is a Network World affiliate.