Clicky

mobile btn
Thursday, March 28th, 2024

IARPA, U.S. Army launch program to rethink computer architecture, aid data analysis

© Shutterstock

Working hand-in-hand, the United States Army’s Army Research Laboratory and the Intelligence Advanced Research Projects Activity (IARPA) seek to innovate computer architecture designs and advanced data analysis through a new program called AGILE.

AGILE, or the Advanced Graphic Intelligence Logical Computing Environment program, is built on the concept of next-generation computing for the Intelligence Community. By deploying contracts to a variety of partners – in this case, Advanced Micro Devices, Inc., The Georgia Tech Research Corporation, Intel Federal LLC, the Trustees of Indiana University, Qualcomm Intelligent Solutions, Inc. and the University of Chicago – this research and development will focus on three areas.

First, it will encourage researchers to break down the traditional concept of computer infrastructure and create new mechanisms for the access, movement and storage of complex data streams and the structures that allow execution of data-analytic algorithms. Second, teams will advance the field of predictive analysis, working to speed up the processing of massive data troves. With all that will be generated, the government also hopes this will lead to significant performance increases for the Intelligence Community.

“The start of this research means that we are now on the path to design computers that can process exponentially increasing amounts and sources of data,” Dr. William Harrod, program manager, said. “Using outside-the-box thinking, we aim to design computer architectures to address today’s data-intensive problems while generating important analytical insights in a timely way.”

In addition to the contractors for the design side, AGILE also includes a test and evaluation team. Lawrence Berkeley National Laboratory, Sandia National Laboratory and Pacific Northwest National Laboratory fill that side of things, thanks to their experience in evaluating computer architecture design and performance.