Center for Magnetic Recording Research

Research Interests

The Non-volatile Systems Laboratory

Steven Swanson is a highly respected experimental computer architect. While working on his Ph.D. at the University of Washington, he developed WaveScalar, a dataflow processor architecture that avoids the scaling and complexity problems of conventional von Neumann processors, while solving the dataflow memory ordering problem of previous dataflow machines. As a result, dataflow processing is now a viable alternative to the von Neumann execution that dominates currently available processors. WaveScalar's ability to extract parallelism at many levels within a single application while supporting familiar memory semantics provides an opportunity to accelerate computer performance independent of underlying technology scaling trends. In developing WaveScalar, professor Swanson also developed new techniques that allow programmers to express sophisticated information to the processor, allowing it to perform some operations 100s of times faster than conventional systems. Currently, he is developing techniques for refining the architecture's interactions with programming languages, software engineering, and technology constraints to make it easier to build high-performance, power-efficient computing systems that are scalable and easy to maintain.

Biography

Steven Swanson is an assistant professor of Computer Science and Engineering at UCSD's Jacobs School of Engineering. He received his Ph.D. in Computer Science and Engineering from the University of Washington in 2006 and a Bachelor of Science in Computer Science and Mathematics from the University of Puget Sound in 1999. At UCSD, Swanson teaches undergraduate and graduate courses in computer architecture, machine organization, and hardware/software interfaces. He also trains graduate students in computer architecture with emphasis on its interaction with software system components. Swanson received graduate research fellowships from both the National Science Foundation (1999) and Intel (2002).

Publications

CMRR Related Publications

1. Grupp, L., Caulfield, A.M., Coburn, J., Yaakobi, E., Swanson, S., Siegel, P., "Characterizing Flash Memory: Anomalies, Observations, and Applications" To appear in MICRO'09.  (pdf) (Flash memory summit slides) (More information about our flash characterization project is available here.)

Student Name:

Interests:

email:

Phone:

Fax:

Office:

Mail Code:

Graduate Students

Student Name:

Interests:

email:

Phone:

Fax:

Office: