Revealing Untapped DSP Optimization Potentials for FPGA-Based Systolic Matrix Engines
Authors:
- 2 September 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 197-203
Abstract
Systolic architectures are widely embraced by neural network accelerators for their superior performance in highly parallelized computation. The DSP48E2s serve as dedicated arithmetic blocks in Xilinx Ultrascale series FPGAs and constitute a fundamental component in FPGA-based systolic matrix engines. Harnessing the full potential of DSP48E2s in architectural design can result in significant performance enhancements for systolic architectures on Ultrascale series FPGAs. This paper unveils several previously untapped DSP optimization techniques capable of further enhancing FPGA-based systolic matrix engines. We apply these techniques to two well-known systolic architectures: Google TPUv1 and Xilinx Vitis AI DPU. With the proposed techniques, our design achieves substantial resource and power reduction compared to the open-source TPUv1 FPGA implementation and the Vitis AI DPU implementation in the same parallelism setting. We also demonstrate the applicability of our techniques to neuromorphic hardware for supporting spiking neural network acceleration.Keywords
All Related Versions
Funding Information
- Chinese Academy of Sciences
This publication has 15 references indexed in Scilit:
- FireFly v2: Advancing Hardware Support for High-Performance Spiking Neural Network With a Spatiotemporal FPGA AcceleratorIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2024
- Uint-Packing: Multiply Your DNN Accelerator Performance via Unsigned Integer DSP Packing2023
- BrainCog: A spiking neural network based, brain-inspired cognitive intelligence engine for brain-inspired AI and brain simulationPatterns, 2023
- FireFly: A High-Throughput Hardware Accelerator for Spiking Neural Networks With Efficient DSP and Memory OptimizationIEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2023
- Efficient Error Detection for Matrix Multiplication With Systolic Arrays on FPGAsIEEE Transactions on Computers, 2023
- BP-Im2col: Implicit Im2col Supporting AI Backpropagation on Systolic Arrays2022
- DSP-Packing: Squeezing Low-precision Arithmetic into FPGA DSP Blocks2022
- Hardware Resource and Computational Density Efficient CNN Accelerator Design Based on FPGA2021
- Scaling the Cascades: Interconnect-Aware FPGA Implementation of Machine Learning Problems2019
- Double MAC on a DSP: Boosting the Performance of Convolutional Neural Networks on FPGAsIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2018