Field-Programmable Gate Array Architecture for Deep Learning: Survey and Future Directions
Open Access
- 30 October 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Proceedings of the IEEE
- Vol. 113 (7) , 613-639
Abstract
Deep learning (DL) is becoming the cornerstone of numerous applications both in large-scale datacenters and at the edge. Specialized hardware is often necessary to meet the performance requirements of state-of-the-art DL models, but the rapid pace of change in DL models and the wide variety of systems integrating DL make it impossible to create custom computer chips for all but the largest markets. Field-programmable gate arrays (FPGAs) present a unique blend of reprogrammability and direct hardware execution that make them suitable for accelerating DL inference. They offer the ability to customize processing pipelines and memory hierarchies to achieve lower latency and higher energy efficiency compared to general-purpose central processing units (CPUs) and graphics processing units (GPUs), at a fraction of the development time and cost of custom chips. Their diverse and high-speed inputs/outputs (IOs) also enable directly interfacing the FPGA to the network and/or a variety of external sensors, making them suitable for both datacenter and edge use cases. As DL has become an ever more important workload, FPGA architectures are evolving to enable higher DL performance. In this article, we survey both academic and industrial FPGA chip architecture enhancements for DL. First, we give a brief introduction on the basics of FPGA architecture and how its components lead to strengths and weaknesses for DL applications. Next, we discuss different design styles of DL inference accelerators implemented on FPGAs that achieve state-of-the-art performance and productive development flows, ranging from model-specific dataflow styles to software-programmable overlay styles. We survey DL-specific enhancements to traditional FPGA building blocks including the logic blocks (LBs), arithmetic circuitry, and on-chip memories, as well as new DL-specialized blocks that integrate into the FPGA fabric to accelerate tensor computations. Finally, we discuss hybrid devices that combine processors and coarse-grained accelerator blocks with FPGA-like interconnect and networks-on-chip (NoCs), and highlight promising future research directions.Keywords
All Related Versions
Funding Information
- Natural Sciences and Engineering Research Council of Canada
- United States National Science Foundation (2417658)
- Intel/VMWare Crossroads 3D-FPGA Academic Research Center, and Advanced Micro Devices, Inc
This publication has 118 references indexed in Scilit:
- Discovering faster matrix multiplication algorithms with reinforcement learningNature, 2022
- Architecture and Application Co-Design for Beyond-FPGA Reconfigurable Acceleration DevicesIEEE Access, 2022
- Highly accurate protein structure prediction with AlphaFoldNature, 2021
- How to stop data centres from gobbling up the world’s electricityNature, 2018
- CGRA-ME: A unified framework for CGRA modelling and exploration2017
- Embedded Multi-die Interconnect Bridge (EMIB) -- A High Density, High Bandwidth Packaging Interconnect2016
- FPGA Acceleration of Recurrent Neural Network Based Language Model2015
- ABC: An Academic Industrial-Strength Verification Tool2010
- Odin II - An Open-Source Verilog HDL Synthesis Tool for CAD Research2010
- Histograms of Oriented Gradients for Human Detection2005