Todd Austin recognized for outstanding achievements

Prof. Todd Austin has been recognized with the University of Michigan’s Distinguished Faculty Achievement Award, which honors senior faculty who have consistently demonstrated outstanding achievements in the areas of scholarly research or creative endeavors, teaching and mentoring of students and junior faculty, service, and a variety of other activities. Up to five awards of $2,500 are made each year.

Austin is a recognized leader in the area of computer architecture research, having delivered several seminal papers that have changed the landscape of research in the field. He is the 12th and 13th most cited author on the MICRO and ISCA Hall of Fame, respectively, and a former leader C-FAR, which was the field’s largest research center at the time.

Most recently, Austin began work on with a $3.6M DARPA grant to design an “unhackable” computer that is inspired by the human immune system. He leads the project, called Morpheus, which uses an immunological approach that is dramatically different from today’s software approaches. The project outlines a new way to design hardware so that information is rapidly and randomly moved and destroyed. The technology works to elude the attacker from the critical information they need to construct a successful attack—it could protect both hardware and software.

In 2016, Austin led a project that identified a hardware backdoor that can be secretly added to a microprocessor design late in the fabrication stage. This backdoor could be triggered later via software commands to trip the system’s “privilege” bit and take over control of the host system. His paper, “A2: Analog Malicious Hardware,” earned top honors at the 37th IEEE Symposium on Security and Privacy.

Austin’s past influential projects include Bulletproof, a set of novel, ultra low-cost mechanisms to protect a microprocessor pipeline and on-chip memory system from silicon defects, and Razor, a low-power processor design based on timing speculation that increases energy efficiency with little performance impact. The latter inspired the movement for “probabilistic computing” in which an architecture is permissive of errors in order to save power, time, and other resources. In earlier work, he released SimpleScalar, one of the first execution-driven architectural simulators, and his paper describing it has been cited over 3500 times. The simulator led to a revolution of quantitative computer architecture research where performance analysis became a required ingredient of every published paper.

In 2017, he was named an IEEE Fellow for his contributions to simulation techniques and resilient system design in computer architecture. Amongst his many recognitions, he has twice received the Richard Newton Gigascale Systems Research Center Industrial Impact Award, the ACM/SIGARCH Maurice Wilkes Award, the U-M Henry Russel Award, and a number of research and teaching awards from the College of Engineering. Early in his career at Michigan, he received an NSF CAREER Award and was selected as a Sloan Fellow.