Stencil-HMLS: A multi-layered approach to the automatic optimisation of stencil codes on FPGA

Gabriel Rodriguez-Canal, Nick Brown, Maurice Jamieson, Emilien Bauer, Anton Lydike, Tobias Grosser

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

The challenges associated with effectively programming FPGAs have been a major blocker in popularising reconfigurable architectures for HPC workloads. However new compiler technologies, such as MLIR, are providing new capabilities which potentially deliver the ability to extract domain specific information and drive automatic structuring of codes for FPGAs. In this paper we explore domain specific optimisations for stencils, a fundamental access pattern in scientific computing, to obtain high performance on FPGAs via automated code structuring. We propose Stencil-HMLS, a multi-layered approach to automatic optimisation of stencil codes and introduce the HLS dialect, which brings FPGA programming into the MLIR ecosystem. Using the PSyclone Fortran DSL, we demonstrate an improvement of 14-100 × with respect to the next best performant state-of-the-art tool. Furthermore, our approach is 14-92 × more energy efficient than the next most energy efficient approach.

Original languageEnglish
Title of host publicationProceedings of 2023 SC Workshops of the International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023
PublisherAssociation for Computing Machinery
Pages556-565
Number of pages10
ISBN (Electronic)9798400707858
DOIs
Publication statusPublished - 12 Nov 2023
Event2023 International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023 - Denver, United States
Duration: 12 Nov 202317 Nov 2023

Publication series

NameACM International Conference Proceeding Series

Conference

Conference2023 International Conference on High Performance Computing, Network, Storage, and Analysis, SC Workshops 2023
Country/TerritoryUnited States
CityDenver
Period12/11/2317/11/23

Keywords / Materials (for Non-textual outputs)

  • FPGAs
  • HPC
  • MLIR
  • stencil based codes
  • U280
  • xDSL

Fingerprint

Dive into the research topics of 'Stencil-HMLS: A multi-layered approach to the automatic optimisation of stencil codes on FPGA'. Together they form a unique fingerprint.

Cite this