Harnessing FPGAs potential with OpenCL

Abstract : The work presented deals with the evaluation of F-PGAs resurgence for hardware and software acceleration. We focus our attention on the tools developed by FPGAs manufacturers, in particular the Intel FPGA S-DK for OpenCL, that promises a new level of hardware abstraction from the developer's perspective, allowing a software-like programming of FPGAs. Our first contribution is to propose an accurate memory benchmark, and we follow with an evaluation of different custom OpenCL implementations of one use case : the computed tomogra-phy. With some clues on memory fetching and coalescing, we then further tune designs to improve performance. Finally , a comparison is made with GPU implementations, and a preliminary conclusion is drawn on FPGAs future in the semiconductor realm. This presentation include but is not limited to results presented in [3].
Document type :
Conference papers
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01841029
Contributor : Nicolas Gac <>
Submitted on : Tuesday, July 17, 2018 - 8:22:15 AM
Last modification on : Thursday, February 7, 2019 - 3:42:32 PM
Document(s) archivé(s) le : Thursday, October 18, 2018 - 4:01:08 PM

Identifiers

  • HAL Id : hal-01841029, version 1

Citation

Maxime Martelli, Nicolas Gac, Alain Mérigot, Cyrille Enderli. Harnessing FPGAs potential with OpenCL. 13ème Colloque nationale du GDR SoC/SiP, Jun 2018, Paris, France. ⟨hal-01841029⟩

Share

Metrics

Record views

204

Files downloads

73