IBM gives grid computing a push
Thursday, August 2, 2001
By Jeffrey Burt, eWEEK
August 2, 2001 3:06 PM ET
John Patrick has been involved in high tech and the Internet for years.
“I’ve often been asked, ‘What’s the next big thing for the Internet?'” said Patrick, IBM’s vice president for Internet strategies. “Until now, I didn’t have the answer. I’m very confident now that the next big thing will be grid computing.”
Grid computing is a step up from distributed computing, according to Patrick. Primarily aimed at universities and researchers, but with business possibilities as well, grid computing is a way of bringing together supercomputers, servers and storage devices via open standards to power very large computing jobs.
IBM is pushing to become a key technology player behind the growth of grid computing.
To that end, Big Blue today announced it has been selected by the British government to form the bulk of the technology behind the creation of the National Grid in the United Kingdom, which will link nine grid centers at universities. It is working on a similar project in the Netherlands.
The company also said it will work with the open-source groups Global Grid Forum and the Globus development community to develop open grid standards, and it is creating its own grid computing initiative that will allow customers to tap into IBM’s vast computing capabilities when needed.
SETI writ large
Patrick said grid computing is similar to initiatives such as [email protected], a peer-to-peer effort in which people allow their computers to be used to search for radio signals from outer space when they’re not using the machines. But while those projects use desktop PCs, the idea behind grid computing is linking supercomputers, applications, servers and storage technology.
However, like those initiatives, grid computing primarily targets universities and institutions whose researchers are running projects that call for enormous computing capabilities. For example, Patrick said medical companies could create cures to diseases specifically targeting individual patients or track diseases across continents to determine where in the world to move the medicine to fight the disease.
But it also will have implications in the business world — in massive, large-scale hosting environments, for example, he said. Collaboration and database sharing are other possibilities.
Patrick said IBM’s Project eLiza, announced in April, also will be an important technology. The multibillion-dollar initiative calls for creating self-managing servers that require little or new human interaction.
The idea of grid computing isn’t new, but the technology is beginning to catch up with the theory.
“It’s time for it to become real now,” Patrick said. “We’re moving from the design phase to the implementation phase.”
IBM is not the only company moving on this front. Last week at the O’Reilly open-source conference in San Diego, Sun Microsystems Inc. announced the Grid Engine Project, which includes software that locates idle computing resources, matches them to individual jobs and brings the computing power back to the desktop.
Sun has made the more than half-million lines of code for the software available for download at gridengine.sourcenet.com.