Publication Details

HLS-based Fault Tolerance Approach for SRAM-based FPGAs

LOJDA Jakub, PODIVÍNSKÝ Jakub, KRČMA Martin and KOTÁSEK Zdeněk. HLS-based Fault Tolerance Approach for SRAM-based FPGAs. In: Proceedings of the 2016 International Conference on Field Programmable Technology. Xi'an: IEEE Computer Society, 2016, pp. 301-302. ISBN 978-1-5090-5602-6.
Czech title
Zajištění odolnosti proti poruchám na úrovni HLS pro FPGA založená na SRAM
Type
conference paper
Language
english
Authors
Keywords

High Level Synthesis
CatapultC
Fault Tolerance
Robot Controller

Abstract

This paper presents an approach to fault-tolerant systems design and synthesis based on High-level Synthesis (HLS). A description and evaluation of the impacts of HLS optimization methods are shown as well. The higher reliability is achieved through modification of input description in the C++ programming language on which the HLS synthesis tools are based on. Our work targets SRAM-based FPGAs, which are prone to Single Event Upsets (SEUs). For the evaluation of impacts of HLS optimization methods we use our evaluation platform, which allows us to test fault tolerance properties of the Design Under Test (DUT). The evaluation platform is based on functional verification combined with fault injection.

Published
2016
Pages
301-302
Proceedings
Proceedings of the 2016 International Conference on Field Programmable Technology
Conference
The 2016 International Conference on Field-Programmable Technology, Xi'an, CN
ISBN
978-1-5090-5602-6
Publisher
IEEE Computer Society
Place
Xi'an, CN
DOI
UT WoS
000402988900057
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB11275,
   author = "Jakub Lojda and Jakub Podiv\'{i}nsk\'{y} and Martin Kr\v{c}ma and Zden\v{e}k Kot\'{a}sek",
   title = "HLS-based Fault Tolerance Approach for SRAM-based FPGAs",
   pages = "301--302",
   booktitle = "Proceedings of the 2016 International Conference on Field Programmable Technology",
   year = 2016,
   location = "Xi'an, CN",
   publisher = "IEEE Computer Society",
   ISBN = "978-1-5090-5602-6",
   doi = "10.1109/FPT.2016.7929561",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/11275"
}
Back to top