The research, led by Assistant Professor Kevin Angstadt ’14, was recently published in the September/October 2022 Special Issue of IEEE Micro, a journal of accelerators.
The demands of business leaders for real-time data analysis and the technical challenges of designing faster general-purpose processors have led to the development and adoption of specialized, application-specific computer hardware known as accelerators.
Angstadt paper, Integrating legacy line code for FPGAs using Bounded Automata Learning, which focuses on automatically converting existing software to run on these new accelerators. Existing programming techniques require specialized training and expertise, but the new method presented by Angstadt and colleagues can significantly reduce this burden for certain types of applications in many domains, such as virus scanning, network security, social network analysis, machine learning, and bioinformatics. . paper published in IEEE Micro It also includes test results showing that Angstadt’s new method improves the speed of released programs and reduces hardware requirements compared to current industry standard tools.
This article is a collaboration between Kevin Angstadt of St. Lawrence University, Tommy Tracy II and Kevin Skadron of the University of Virginia, and Jean-Baptiste Jannin and Westley Weimer of the University of Michigan. This work was supported in part by the National Science Foundation, the Air Force Research Laboratory, the Jefferson Scholars Foundation, and the Center for Research in Intelligent Storage and Processing in Memory (CRISP), one of six centers in the Joint University Microelectronics Program (JUMP). , a Semiconductor Research Corporation (SRC) program sponsored by the Defense Advanced Research Projects Agency (DARPA).