IBM Picks Brains in Tucson

IBM picks U of Arizona to research strategic storage networking technologies

August 11, 2004

3 Min Read
Network Computing logo

Desert heat could bring new SAN technologies to the boil at the University of Arizona in Tucson, thanks to an IBM Corp. (NYSE: IBM) research grant.

Like most big companies, IBM enlists university researchers for R&D. In this case, it's bequeathing a raft of hardware for projects focused on two key areas crucial to its storage strategy -- dynamic provisioning and storage resource management.

IBM picked the university's business school, the Eller College of Management, for one of the company's Shared University Research grants. "They have a unique blend of technology and business knowledge. That's one of their strengths," says Ken Boyd, distinguished engineer at IBM.

IBM wants two things from this project: First, it wants the researchers to find better ways to automatically assign storage to users -- regardless of where they are located. In most SANs today, users can tap into the SAN only when they're in a specific location. IBM hopes to better accommodate users who work outside the office, make requests from different locations, or need different data from geographically dispersed servers.

"The way storage needs to be allocated today is constantly changing -- it's no longer static," says Dr. Sudha Ram, Eller professor of MIS (management information systems) at the University of Arizona and the lead investigator for the research project. Ram has done similar research in data management and system interoperability in the past. IBM's Boyd indicates the project could lead to new designs for SAN gear, such as integrating virtual-resource allocation into storage controllers.Second, Ram also plans to study the issue of automatically reassigning storage that is no longer required for specific data -- a promising avenue of research for IBM, given the trend toward all things ILM (information lifecycle management).

According to IBM's Ken Boyd, current storage products don't make the most of reclaiming storage resources once they're freed up. New techniques could improve SAN performance and efficiency. He says the researchers will "probe deeper into the question of how you create and go after unused storage."

The research dovetails with IBM's "on demand" computing strategy, but it also appeals to Big Blue for other reasons. It will include wireless technologies, which are becoming a more prevalent means of making data-access requests. Ram and her team also will study the administrative aspects of data allocation, such as streamlining the steps involved in responding to requests.

If the project goes as planned, IBM will get some key R&D for next to no cost: The grant includes four IBM eServer xSeries 325 systems, one IBM eServer xSeries 445 system, six IBM ThinkPad notebook computers with wireless capabilities, and two IBM IntelliStation workstations. The equipment will join a previously received IBM LTO Library and an IBM TotalStorage FAStT Storage Server for the upcoming research.

Researchers will get bragging rights. The university will get the prestige of working with IBM -- and the rights to publish their conclusions for general use. What end users will get depends on how IBM and others implement what the Tucson team finds.Brett Mendel, Senior Analyst, Byte and Switch Insider

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like


More Insights