GrammaTech, Inc., a leading provider of source-code analysis tools, announced that it has been awarded a DARPA contract to develop static-analysis technology for examining multi-core systems. The new technology focuses on identifying concurrent programming defects, which are notoriously difficult to identify and manually debug.
Multi-core processors are integrated circuits that contain two or more processor cores. Well-known examples include Intel’s Core 2 Duo, AMD’s Phenom, and Sony, Toshiba and IBM’s (STI’s) Cell Processor. The adoption of multi-core is being fueled by the need for better performance in an age when improvements to the single-processor architecture have already been so dramatic that future improvements are being hampered by the laws of physics. Multi-core processors provide an avenue for improving performance through parallel computation, and they are widely used in multimedia applications, scientific applications and CAD/CAM packages.
Research recently released by Venture Development Corporation (VDC), projects strong revenue growth through 2011 in the market for multi-core CPUs in embedded applications. According to Eric Heikkila, director of VDC’s Embedded Hardware and Software Practice, “The embedded multi-core CPU market very much equates to a billion dollar opportunity over the next five years.”
“The use of multi-core chips is growing, but it is difficult to maximize the full performance of multi-core technology. Software for multi-core processors must be explicitly parallel, and it must account for the low-level details of how the cores pass data between each other and memory. Static analysis can help developers by identifying errors that commonly occur when programming for a multi-core system. Our goal is to provide a static analysis technology that helps developers avoid the increases in development cost, despite the increase in complexity,” said David Melski, VP of Research at GrammaTech.
GrammaTech’s static-analysis tools are used worldwide by startups, Fortune 500 companies, educational institutions, and government agencies. The staff includes ten researchers with PhDs in programming languages and program analysis. The company has offices in Ithaca, New York, and San Jose, California.