License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/OASIcs.PARMA-DITAM.2022.6
URN: urn:nbn:de:0030-drops-161225
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2022/16122/
Go to the corresponding OASIcs Volume Portal

Ratto, Francesco ; Esposito, Stefano ; Sau, Carlo ; Raffo, Luigi ; Palumbo, Francesca

Multithread Accelerators on FPGAs: A Dataflow-Based Approach

pdf-format:
OASIcs-PARMA-DITAM-2022-6.pdf (4 MB)

Abstract

Multithreading is a well-known technique for general-purpose systems to deliver a substantial performance gain, raising resource efficiency by exploiting underutilization periods. With the increase of specialized hardware, resource efficiency became fundamental to master the introduced overhead of such kind of devices. In this work, we propose a model-based approach for designing specialized multithread hardware accelerators. This novel approach exploits dataflow models of applications and tagged tokens to let the resulting hardware support concurrent threads without the need to replicate the whole accelerator. Assessment is carried out over different versions of an accelerator for a compute-intensive step of modern video coding algorithms, under several feeding configurations. Results highlight that the proposed multithread accelerators achieve a valuable tradeoff: saving computational resources with respect to replicated parallel single-thread accelerators, while guaranteeing shorter waiting, response, and elaboration time than a unique single-thread accelerator multiplexed in time.

BibTeX - Entry

@InProceedings{ratto_et_al:OASIcs.PARMA-DITAM.2022.6,
  author =	{Ratto, Francesco and Esposito, Stefano and Sau, Carlo and Raffo, Luigi and Palumbo, Francesca},
  title =	{{Multithread Accelerators on FPGAs: A Dataflow-Based Approach}},
  booktitle =	{13th Workshop on Parallel Programming and Run-Time Management Techniques for Many-Core Architectures and 11th Workshop on Design Tools and Architectures for Multicore Embedded Computing Platforms (PARMA-DITAM 2022)},
  pages =	{6:1--6:14},
  series =	{Open Access Series in Informatics (OASIcs)},
  ISBN =	{978-3-95977-231-0},
  ISSN =	{2190-6807},
  year =	{2022},
  volume =	{100},
  editor =	{Palumbo, Francesca and Bispo, Jo\~{a}o and Cherubin, Stefano},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2022/16122},
  URN =		{urn:nbn:de:0030-drops-161225},
  doi =		{10.4230/OASIcs.PARMA-DITAM.2022.6},
  annote =	{Keywords: multithreading, dataflow, hardware acceleration, heterogeneous systems, tagged dataflow}
}

Keywords: multithreading, dataflow, hardware acceleration, heterogeneous systems, tagged dataflow
Collection: 13th Workshop on Parallel Programming and Run-Time Management Techniques for Many-Core Architectures and 11th Workshop on Design Tools and Architectures for Multicore Embedded Computing Platforms (PARMA-DITAM 2022)
Issue Date: 2022
Date of publication: 08.06.2022

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI