Portrait of Susan Eggers in a suit standing outside in front of a brick wall framed by foliageSusan Eggers joined the University of Washington faculty in 1989. A leading computer architect, she was elected a Member of the National Academy of Engineering in 2006 and a Fellow of the American Academy of Arts & Sciences in 2013. She was also elected a Fellow of ACM, IEEE, and the American Association for the Advancement of Science. In 2018 she became the first woman to receive the Eckert-Mauchly Award, the most prestigious honor in the computer architecture field. She retired in 2011.

After earning her bachelor’s degree in economics, Susan worked in a variety of roles for nearly two decades before turning her attention to computer architecture. She joined the University of Washington at the age of 47 after completing her Ph.D. at University of California, Berkeley. Susan would more than make up for lost time by producing some of the most significant and enduring contributions to the field of computer architecture.

Most noteworthy among these was her role in the development and commercialization of simultaneous multithreading (SMT). While chip manufacturers were achieving rapid gains in memory and logic in the mid-1990s, those physical manifestations of Moore’s Law failed to generate the expected improvements in performance. To Susan, the most promising approach to translate exponential growth in chip density into enhanced performance was to increase parallelism, or the ability of computers to run calculations concurrently.

Susan — working with her Allen School faculty colleague Hank Levy, several graduate students, and Joel Emer (then at Digital Equipment Corporation) — drove the SMT revolution, which boosted parallelism, and with it, performance. SMT combines hardware multithreading with superscalar processor technology to enable multiple independent threads to issue instructions to multiple functional units in a single cycle. Susan and her collaborators demonstrated several substantial advantages that SMT offered over other architectures, including higher throughput, increased speed, and greater flexibility in hardware design. SMT faced considerable skepticism, but over the next eight years Susan and her collaborators in academia and industry would refine and validate SMT, which became an essential component in the processors produced by industry leaders such as Intel and IBM. Susan co-authored roughly a dozen papers about SMT, two of which earned Test of Time Awards (in 2010 and 2011, respectively) from the International Symposium on Computer Architecture.

Susan also made significant, early-career contributions in cache coherency, a technique for maintaining consistent data across shared memory multiprocessors. These included the first data-driven study of multiprocessor data sharing — which was instrumental in advancing the field’s understanding of hardware and software coherency techniques — as well as novel cache coherency protocols.

Susan participated in WaveScalar, a dataflow instruction set architecture and execution model designed for scalable, low-complexity/high-performance processors. WaveScalar is unique among dataflow architectures in efficiently providing traditional memory semantics to execute applications written in imperative, rather than hardware description, languages.

Susan's interests in performance improvements extended beyond chip design. For example, she was a member of the team that built DyC, an easy-to-use system for dynamic compilation in C that was more expressive, flexible, and controllable than previous annotation-based approaches. She was also a member of the team that developed CHiMPS, a C-to-FPGA synthesis compiler that enabled developers to program FPGAs in an imperative language and memory model, while still providing greater performance and less power than a CPU.

A wonderful article describing Susan's extraordinary journey was published by the IEEE Computer Society at the time of her receipt of the Eckert-Mauchly Award. Susan concluded her inspirational acceptance speech with a comment on retirement life: "Although I’m no longer a computer architect, I haven’t left architecture. What you see behind me are the results of a new in-retirement career as a landscape architect. I’ve transformed about half an acre of back yard from a steep slope of blackberries, horse tails, and bog into a lovely garden of small trees, shrubs, perennials, and groundcover with winding stone paths up the hillside and a small stream descending down. I love it! Retirement is pretty great too."

Holders of the Susan J. Eggers Professorship

  • Michael Ernst, 2022–Present