Users optimize capacity with grids, Part 1

Grid computing can help enterprise IT managers achieve better performance without investing in more hardware, say leading-edge users who shared their early-adopter stories with Comdex trade show attendees last month.

The three speakers detailed how they work with the Globus Alliance and its Globus Toolkit, an open-source implementation of the Open Grid Services Infrastructure (OGSI) specification.

At its core, the Globus Alliance is a research project. Globus research focuses on the issues associated with building computational grid infrastructures and on developing applications that use grid services. OGSI provides a common framework for grid services, so developing new OGSI-compliant services is straightforward, according to the Globus Alliance. Every OGSI-compliant service can be used and managed via common interfaces, so building systems and applications with OGSI-compliant services can be easier.

Suzanne George, director of operations and IT at GlobeXplorer, a provider of satellite images and aerial photography via the Internet, said that to enable her network infrastructure to deliver high-resolution photos and maps in seconds, she needed to roll out a Globus-enabled application.

With weather sites and online companies such as MapQuest depending on GlobeXplorer to be able to update images when end users click on a Web page, George said she needed to put a process in place that would allow reliable and ubiquitous access; enable metered delivery of a uniform product; and equip her infrastructure to ingest more than 200 terabytes of raw data in less than four months. And the solution could not require purchasing any new hardware.

“With the grid application, I can scale it across as many platforms as I need to,” George said. The grid application let George use resources at GlobeXplorer’s remote sites and process images off-site, which kept the on-site resources operating smoothly. The Globus grid application tapped unused capacity at remote locations and allowed George to do more with less. “We were able to ingest quickly more content for less money,” she said.

Next time: How others implemented grid computing.