Bayesian Optimization for Efficient Accelerator Synthesis

Journal Article (Journal Article)

Accelerator design is expensive due to the effort required to understand an algorithm and optimize the design. Architects have embraced two technologies to reduce costs. High-level synthesis automatically generates hardware from code. Reconfigurable fabrics instantiate accelerators while avoiding fabrication costs for custom circuits. We further reduce design effort with statistical learning. We build an automated framework, called Prospector, that uses Bayesian techniques to optimize synthesis directives, reducing execution latency and resource usage in field-programmable gate arrays. We show in a certain amount of time that designs discovered by Prospector are closer to Pareto-efficient designs compared to prior approaches. Prospector permits new studies for heterogeneous accelerators.

Full Text

Duke Authors

Cited Authors

  • Mehrabi, A; Manocha, A; Lee, BC; Sorin, DJ

Published Date

  • January 1, 2021

Published In

Volume / Issue

  • 18 / 1

Electronic International Standard Serial Number (EISSN)

  • 1544-3973

International Standard Serial Number (ISSN)

  • 1544-3566

Digital Object Identifier (DOI)

  • 10.1145/3427377

Citation Source

  • Scopus