NSF funds up a storm on grids

The National Science Foundation is funding a grid computing effort aimed at helping scientists to more accurately forecast weather conditions.

On Wednesday, the NSF announced a $11.25 million IT research grant to support a project called Linked Environments for Atmospheric Discovery (LEAD), in which the National Center for Supercomputing Applications (NCSA) will collaborate with eight U.S. academic institutions.

The NSF said the project will help researchers run atmospheric models in much more realistic and real-time settings than is now possible. It intends to develop grid computing tools for on-demand detection, simulation and prediction of weather events such as thunderstorms and tornadoes.

Under the project, researchers will get access to a broad array of tools and to databases from their desktops. Data could be stored at distributed sites and used to simulate models, then finally moved to make real-time forecasts. The system should enable identification of thunderstorms as they form, automatically triggering data-gathering tools, requesting grid computing resources, and generating results as the weather unfolds.

"Our ultimate goal is to create a system that takes full advantage of all the atmospheric data that is constantly being collected, the power of supercomputers, and the speed of high-performance networks," Bob Wilhelmson, a senior research scientist at NCSA, said in a statement.

The NCSA plans to integrate various components of the LEAD project developed at different institutions. The grid and Web applications will be tested at five centers and introduced in three phases over five years.

Another $8.25 NSF million grant is earmarked for a NCSA effort called "v-grads," which targets virtual grid application development software to simplify and accelerate grid applications and services. The LEADS project is part of the v-grads effort.