About Me
I’m a member of the embedded team at Efficient Computer. I received my PhD in Computer Science and BS degrees in Electrical and Computer Engineering from Virginia Tech in 2024 and 2019, respectively. My dissertation explored circuit designs for power management in batteryless energy harvesting systems. As an undergrad, I built systems for detecting counterfeit recycled electronic devices based on aging mechanisms in SRAM (see the paper here).
Publications
A Software Caching Runtime for Embedded NVRAM Systems.
Harrison Williams and Matthew Hicks. ASPLOS 2025.
[PDF]
Overcoming NVRAM access penalties on commercial microcontrollers by shifting code execution into SRAM. This paper technically “appears” in ASPLOS ‘24 Volume 4, which is presented at ASPLOS ‘25.
A Survey of Prototyping Platforms for Intermittent Computing Research.
Harrison Williams and Matthew Hicks. ENSsys 2024.
[PDF]
Best Paper Award.
An exploration of the commercial prototyping platforms available for intermittent system development, how their specific features and parameters affect research conclusions, and some recommended future directions.
A Difference World: High-performance, NVM-invariant, Software-only Intermittent Computation.
Harrison Williams*, Saim Ahmad*, and Matthew Hicks. *Equal contribution. USENIX ATC 2024.
[PDF]
[Slides]
Efficient state rollback and re-execution for SRAM-based intermittent systems.
Energy-Adaptive Buffering for Efficient, Responsive, and Persistent Batteryless Systems.
Harrison Williams and Matthew Hicks. ASPLOS 2024.
[PDF]
[Slides]
Unified, variable-capacitance energy buffer for batteryless systems.
RF Energy Harvesting in Minimization of Age of Information with Updating Erasures.
Fariborz Lohrabi Pour, Harrison Williams, Matthew Hicks, Dong Sam Ha. ISCAS 2023.
[PDF]
Failure Sentinels: Ubiquitous Just-in-time Intermittent Computation via hardware support for continuous, low-cost, fine-grain voltage monitoring.
Harrison Williams, Michael Moukarzel, Matthew Hicks. ISCA 2021.
[PDF]
[Slides]
Low-power, all-digital hardware to monitor energy harvester supply voltage.
Forget Failure: Exploiting SRAM Data Remanence for Low-overhead Intermittent Computation.
Harrison Williams, Xun Jian, Matthew Hicks. ASPLOS 2020.
[PDF]
[Slides]
High-performance intermittent computation without high-performance persistent memory using SRAM data retention.
Teaching
Principles of Computer Security (CS 4264 @ Virginia Tech)
Teaching Assistant - Fall 2019
This course teaches the security mindset and introduces the principles and practices of computer security as applied to software, host systems, and networks. It covers the foundations of building, using, and managing secure systems. Topics include standard cryptographic functions and protocols, threats and defenses for real-world systems, incident response, and computer forensics.
Recognition
Davenport Leadership Scholarship (2022).
NSF Graduate Research Fellowship Program, Honorable Mention (2021).
Contact
harrison@efficient.computer