Siying Feng awarded Rackham Predoctoral Fellowship to accelerate performance of emerging computer architectures

The fellowship will help support her work to develop acceleration techniques for emerging computer architectures in applications including machine learning, graph analytics, and scientific computing.
Siying Feng
Siying Feng

PhD student Siying Feng has received a Rackham Predoctoral Fellowship to support her studies in computer science and engineering at the University of Michigan.

Siying’s work is centered around developing acceleration techniques for emerging computer architectures for applications including machine learning, graph analytics, and scientific computing.

The emergence of big data and massive social networks has led to enormous data processing challenges. At the same time, the ability to improve processing performance through scaling has diminished significantly over the last decade and a half, making it increasingly difficult to build hardware for emerging applications that meet power and performance targets, while remaining flexible and programmable for end users.

Siying’s research is aimed at developing and enabling Transmuter, a “software-defined hardware” architecture under development at Michigan in partnership with University of Edinburgh, Arizona State University, and ARM that can change how programs use the hardware available to them in real time, effectively acting as a reconfigurable computer. Transmuter monitors an application’s behavior and can adapt to new demands, altering how hardware resources are deployed

Siying also designed CoSPARSE, an intelligent software framework which takes advantage of Transmuter to judiciously determine the best software algorithm and hardware configuration during execution for acceleration of diverse graph algorithms. Finally, she enabled  MeNDA, a near-memory multi-way merge accelerator for sparse transposition and dataflows.

Siying’s work has been peer-reviewed and published in conferences in the domains of computer architecture (HPCA, PACT, ISCA, MICRO, DAC), system characterization (IISWC, ISPASS), and circuit design (VLSI). She is a part of the DARPA Software-Defined Hardware (SDH) project, which is a joint effort between the University of Michigan, the University of Edinburgh, Arizona State University, and ARM, Inc.

Siying is advised in her studies by Prof. Ronald Dreslinski and works closely with Bredt Family Professor of Engineering Trevor Mudge in CSE, and Kensall D. Wise Collegiate Professor of Electrical Engineering and Computer Science David Blaauw in ECE. 

About the Rackham Predoctoral Fellowship

The Rackham Predoctoral Fellowship supports outstanding doctoral students who have achieved candidacy and are actively working on dissertation research and writing. They seek to support students working on dissertations that are unusually creative, ambitious and risk-taking.

Chip Design & Architectures; David Blaauw; Graduate students; Honors and Awards; Ronald Dreslinski; Student News; Trevor Mudge