A ter­rorist plants a time bomb along a gas line in a res­i­den­tial neigh­bor­hood. He e-​​mails a photo of the death trap to law enforce­ment offi­cials, but no one can tell exactly where the bomb is located.

The solu­tion may lie in work being done in North­eastern University’s Com­puter Archi­tec­ture Research Lab­o­ra­tory, where elec­trical and com­puter engi­neering pro­fessor David Kaeli and his team have devel­oped super­com­puting hardware/​software tech­nology to pin­point the loca­tion of people, buildings—or even bombs—10 to 15 times faster than tra­di­tional com­puting methods.

North­eastern researchers are col­lab­o­rating on the project with col­leagues at the Uni­ver­sity of Vir­ginia and Advanced Micro Devices (AMD), a Sun­ny­vale, California-​​based com­pany that develops com­puter proces­sors for the com­mer­cial and con­sumer sectors.

The inno­v­a­tive tech­nology, which aligns with Northeastern’s com­mit­ment to research that solves global chal­lenges in health, secu­rity and sus­tain­ability, show­cases the value of using Graphics Pro­cessing Units (GPUs) to help pro­tect the nation’s crit­ical infra­struc­ture, said Kaeli.

GPU tech­nology will have huge impli­ca­tions for the intel­li­gence com­mu­nity,” he said, noting the plan to open-​​source the appli­ca­tion in the near future.

Here’s how it works: A user uploads a photo to an image data­base such as Google Earth. The GPU-​​based appli­ca­tion enables the user to run a lightning-​​fast search, com­paring the photo against images in the data­base until it finds a match.

There’s no appli­ca­tion like it,” said Per­haad Mistry a doc­toral stu­dent in elec­trical and com­puter engi­neering who helped develop the idea while on co-​​op at AMD. “Pin­pointing cer­tain people or places needs to be done quickly.”

Industry leaders have recently rec­og­nized Northeastern’s cutting-​​edge research in this area of supercomputing.

•NVIDIA, a multi­na­tional com­pany that develops graphics pro­cessing units, des­ig­nated Northeastern’s Depart­ment of Elec­trical and Com­puter Engi­neering, in con­junc­tion with Mass­a­chu­setts Gen­eral Hos­pital, as a NVIDIA CUDA Research Center for its ongoing devel­op­ment of GPU-​​accelerated med­ical image analysis tools. (CUDA stands for Com­pute Uni­fied Device Archi­tec­ture, a par­allel com­puting archi­tec­ture devel­oped by NVIDIA.)

•AMD named North­eastern a Strategic Aca­d­emic Research Partner for con­ducting research with the company’s GPUs. AMD also awarded North­eastern a $65,000 grant to con­duct GPU-​​research and invited the North­eastern team to present at Super­com­puting 2010, the largest inter­na­tional con­fer­ence on high per­for­mance com­puting, net­working, storage and analysis.

Kaeli is proud of the part­ner­ship. “AMD is a major employer of North­eastern com­puter engi­neering stu­dents and a leader in the field of GPU com­puting,” he said.