Evolutionary bin packing for memory-efficient dataflow inference acceleration on FPGA

Mairin Kroes, Lucian Petrica, Sorin Cotofana, Michaela Blott

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Convolutional Neural Network (CNN) dataflow inference accelerators implemented in Field-Programmable Gate Arrays (FPGAs) have demonstrated increased energy efficiency and lower latency compared to CNN execution on CPUs or GPUs. However, the complex shapes of CNN parameter memories do not typically map well to FPGA On-Chip Memories (OCM), which results in poor OCM utilization and ultimately limits the size and types of CNNs which can be effectively accelerated on FPGAs. In this work, we present a design methodology that improves the mapping efficiency of CNN parameters to FPGA OCM. We frame the mapping as a bin packing problem and determine that traditional bin packing algorithms are not well suited to solve the problem within FPGA- and CNN-specific constraints. We hybridize genetic algorithms and simulated annealing with traditional bin packing heuristics to create flexible mappers capable of grouping parameter memories such that each group optimally fits FPGA on-chip memories. We evaluate these algorithms on a variety of FPGA inference accelerators. Our hybrid mappers converge to optimal solutions in a matter of seconds for all CNN use-cases, achieve an increase of up to 65% in OCM utilization efficiency for deep CNNs, and are up to 200× faster than current state-of-the-art simulated annealing approaches.

Original languageEnglish
Title of host publicationGECCO 2020 - Proceedings of the 2020 Genetic and Evolutionary Computation Conference
PublisherAssociation for Computing Machinery (ACM)
Pages1125-1133
Number of pages9
ISBN (Electronic)9781450371285
DOIs
Publication statusPublished - 2020
Event2020 Genetic and Evolutionary Computation Conference, GECCO 2020 - Cancun, Mexico
Duration: 8 Jul 202012 Jul 2020

Publication series

NameGECCO 2020 - Proceedings of the 2020 Genetic and Evolutionary Computation Conference

Conference

Conference2020 Genetic and Evolutionary Computation Conference, GECCO 2020
CountryMexico
CityCancun
Period8/07/2012/07/20

Fingerprint Dive into the research topics of 'Evolutionary bin packing for memory-efficient dataflow inference acceleration on FPGA'. Together they form a unique fingerprint.

Cite this