January 18, 2001, 11:10 AM — PROMISING TO DELIVER what company officials are calling "true data sharing," IBM early next year will deliver its Storage Tank, the code name for a suite of storage virtualization software and policy-based management tools.
Linda Sanford, IBM's vice president and group executive for IBM's storage subsystem group, said Storage Tank will be the fix for problems associated with operating heterogeneous storage networks comprised of equipment from mixed vendors, a common industry dilemma.
"When customers talk about interoperable storage networks, they mean not just across a heterogeneous server and operating system, but also heterogeneous storage devices, the disk and tape arrays," Sanford said. "So when a customer says they want a truly interoperable environment, they mean all of that. They want to make sure they can hang off a network not only an IBM Shark but a Compaq modular array system."
Storage Tank will bring unity to all the devices being used in a storage network, regardless of vendor origin, Sanford said.
To do this, Storage Tank technology must be able to "share data at the block level as well as the file level" across each storage device, Sanford said.
"I may need only one piece of data from a certain file, but today I have to get the whole file and strip out what I need," she added.
What Storage Tank will do, according to Sanford, is be able to dip down into the individual block-level data that makes up a file, identify the correct block-level data, understand its format, and translate it correctly to the device that is retrieving the data.
"The real end game is the information utility, providing our customers with information when they need it, wherever it resides on the network," Sanford said.
In a keynote address Tuesday at the Storage Networking World Conference in Orlando, Fla., Sanford said that "for the good of the industry, let's all pledge to drive this industry toward full and true interoperability, across our current platforms, across devices, across software, and across data formats and across networks."
Armonk, N.Y.-based IBM will roll out Storage Tank in phases beginning early next year.