Computerworld

Keeping data warehouses from becoming data graveyards

Researchers receive a US$13.8 grant to build a prototype system for automating data analysis workflows

Scientists have more data than they know what to do with, so University of Southern California researchers are trying to figure out a way to help them make the most of it.

The USC Information Sciences Institute researchers have received US$13.8 million in funding to build a prototype system for automating scientific workflows for climatologists, physicists and others. Such workflows can include thousands of steps.

"Our ability to gather data is surpassing our ability to analyze it," says ISI computer scientist Yolanda Gil, who leads the project, dubbed Windward (more formally, it's called Scaleable Knowledge Discovery through Grid Workflows ). "Our data warehouses are becoming data graveyards."

The project will involve use of artificial intelligence and grid computing. The researchers plan to use grid technology to make artificial intelligence more scalable. One of the researchers, Carl Kesselman, has been honing Globus grid software since the mid-1990s.

One goal is to encourage scientists to build workflow architectures into their research plans so that others can follow them.