One size does not fit all: Implementation trade-offs for iterative stencil computations on FPGAs - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

One size does not fit all: Implementation trade-offs for iterative stencil computations on FPGAs

Résumé

Iterative stencils are kernels in various application domains such as numerical simulations and medical imaging, that merit FPGA acceleration. The best architecture depends on many factors such as the target platform, off-chip memory bandwidth, problem size, and performance requirements. We generate a family of FPGA stencil accelerators targeting emerging System on Chip platforms, (e.g., Xilinx Zynq or Intel SoC). Our designs come with design knobs to explore trade-offs. We also propose performance models to hone in on the most interesting design points, and show how they accurately lead to optimal designs. The optimal choice depends on problem sizes and performance goals.
Fichier principal
Vignette du fichier
StencilsFPL2017.pdf (1003.24 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01655590 , version 1 (11-12-2017)

Identifiants

Citer

Gaël Deest, Tomofumi Yuki, Sanjay Rajopadhye, Steven Derrien. One size does not fit all: Implementation trade-offs for iterative stencil computations on FPGAs. FPL - 27th International Conference on Field Programmable Logic and Applications, Sep 2017, Gand, Belgium. ⟨10.23919/FPL.2017.8056781⟩. ⟨hal-01655590⟩
202 Consultations
227 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More