Publication Details

Xel-FPGAs: An End-to-End Automated Exploration Framework for Approximate Accelerators in FPGA-Based Systems

PRABAKARAN Bharath S., MRÁZEK Vojtěch, VAŠÍČEK Zdeněk, SEKANINA Lukáš and SHAFIQUE Muhammad. Xel-FPGAs: An End-to-End Automated Exploration Framework for Approximate Accelerators in FPGA-Based Systems. In: 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD). San Francisco: Institute of Electrical and Electronics Engineers, 2023, pp. 1-9. ISBN 979-8-3503-1559-2.
Czech title
Xel-FPGA: Koncový automatizovaný framework pro prohledávání aproximovaných akcelerátorů v systémech založených na FPGA
Type
conference paper
Language
english
Authors
Prabakaran Bharath S. (TU-Wien)
Mrázek Vojtěch, Ing., Ph.D. (DCSY FIT BUT)
Vašíček Zdeněk, doc. Ing., Ph.D. (DCSY FIT BUT)
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY FIT BUT)
Shafique Muhammad (TU-Wien)
Keywords

Approximate Computing, Accelerator, FPGA, ASIC, Arithmetic Units, Regression, Models, Synthesis

Abstract

Generation and exploration of approximate circuits and accelerators has been a prominent research domain to achieve energy-efficiency and/or performance improvements.
This research has predominantly focused on ASICs, while not achieving similar gains when deployed for FPGA-based accelerator systems, due to the inherent architectural differences between the two. In this work, we propose a novel framework, Xel-FPGAs, which leverages statistical or machine learning models to effectively explore the architecture-space of state-of-the-art ASIC-based approximate circuits to cater them for FPGA-based systems given a simple RTL description of the target application.
We have also evaluated the scalability of our framework on a multi-stage application using a hierarchical search strategy.
The Xel-FPGAs framework is capable of reducing the exploration time by up to 95%, when compared to the default synthesis, place, and route approaches, while identifying an improved set of Pareto-optimal designs for a given application, when compared to the state-of-the-art. The complete framework is open-source and available online at https://github.com/ehw-fit/xel-fpgas.

Published
2023
Pages
1-9
Proceedings
2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)
Conference
IEEE/ACM International Conference On Computer-Aided Design, San Francisco, California, USA, US
ISBN
979-8-3503-1559-2
Publisher
Institute of Electrical and Electronics Engineers
Place
San Francisco, US
DOI
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB13036,
   author = "S. Bharath Prabakaran and Vojt\v{e}ch Mr\'{a}zek and Zden\v{e}k Va\v{s}\'{i}\v{c}ek and Luk\'{a}\v{s} Sekanina and Muhammad Shafique",
   title = "Xel-FPGAs: An End-to-End Automated Exploration Framework for Approximate Accelerators in FPGA-Based Systems",
   pages = "1--9",
   booktitle = "2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)",
   year = 2023,
   location = "San Francisco, US",
   publisher = "Institute of Electrical and Electronics Engineers",
   ISBN = "979-8-3503-1559-2",
   doi = "10.1109/ICCAD57390.2023.10323678",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/13036"
}
Back to top