Marie Nguyen

My Email Address: marien@andrew.cmu.edu

My resume My CV


Background

I am a PhD student in the ECE department at Carnegie Mellon University. I work with Prof. James Hoe on the broad area of computer architecture.
In particular, my research is on FPGA runtime/overlay using partial reconfiguration (PR), and on FPGA acceleration of real-time vision applications. I am also a member of the Computer Architecture Lab at Carnegie Mellon (CALCM).

Research

Though FPGAs are being increasingly used for compute, they are often used as substitutes for ASICs (tasks are mapped statically on the FPGA and do not change over time). In this research, I am trying to leverage the runtime reprogrammability of FPGAs to be more than just a substitute for ASICs. To support this effort, I have been designing a framework in which multiple real-time vision tasks can spatially and temporally share the FPGA. The framework uses PR (PR allows parts of the FPGA to be reprogrammed at runtime) for enabling quick time-sharing of the FPGA fabric. A prototype framework has been built in the context of the SmartHeadlight project at CMU. Recently, I demonstrated that dynamically managing FPGA resources can result in improved power/energy efficiency compared to a traditional static FPGA solution (where tasks are mapped as fixed functions on the FPGA and do not change over time).

Publications

Quantifying the Benefits of Dynamic Partial Reconfiguration for Embedded Vision Applications (FPL 2019).
Marie Nguyen, Robert Tamburo, Srinivasa Narasimhan and James C. Hoe.
The 29th IEEE International Symposium on Field-Programmable Logic and Applications (FPL 2019).
Time-Shared Execution of Realtime Computer Vision Pipelines by Dynamic Partial Reconfiguration (FPL 2018).
Marie Nguyen and James C. Hoe.
The 28th IEEE International Symposium on Field-Programmable Logic and Applications (FPL 2018).
Amorphous Dynamic Partial Reconfiguration with Flexible Boundaries to Remove Fragmentation (arXiv 2017).
Marie Nguyen and James C. Hoe.
GraphGen: An FPGA Framework for Vertex-Centric Graph Computation (FCCM 2014).
Eriko Nurvitadhi, Gabriel Weisz, Yu Wang, Skand Hurkat, Marie Nguyen, James C. Hoe, Josè F. Martìnez, and Carlos Guestrin.
The 22nd IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM 2014).