Evolving Component Library for Approximate High Level Synthesis

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F16%3APU121663" target="_blank" >RIV/00216305:26230/16:PU121663 - isvavai.cz</a>

Výsledek na webu

<a href="http://www.fit.vutbr.cz/research/pubs/all.php?id=11231" target="_blank" >http://www.fit.vutbr.cz/research/pubs/all.php?id=11231</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/SSCI.2016.7850168" target="_blank" >10.1109/SSCI.2016.7850168</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Evolving Component Library for Approximate High Level Synthesis

Popis výsledku v původním jazyce

An approximate computing approach has recently been introduced for high level circuit synthesis (HLS) in order to make good use of approximate circuits at system and block level. It is assumed in HLS algorithms that a component library containing various implementations of elementary circuit components is available. An open problem is how to construct such a component library in the context of approximate computing, where the component's error is a new design variable and hence many compromise implementations exist for a given component. In this paper, we first introduce a multi-objective Cartesian genetic programming method to create a comprehensive component library containing hundreds of Pareto optimal implementations of approximate 8-bit adders and multipliers, where the error, area and delay are simultaneously optimized. Another multi-objective evolutionary algorithm is employed to solve the so called binding problem of HLS, in which suitable approximate components are assigned to nodes of the data flow graph describing a complex digital circuit. Two approaches are then proposed and compared in order to reduce the size of the library of approximate components.  It is shown that a random subsampling of the component library provides satisfactory results in the context of our study. The proposed methods are evaluated using two benchmark circuits -- the reduce (sum) and DCT circuits.

Název v anglickém jazyce

Evolving Component Library for Approximate High Level Synthesis

Popis výsledku anglicky

An approximate computing approach has recently been introduced for high level circuit synthesis (HLS) in order to make good use of approximate circuits at system and block level. It is assumed in HLS algorithms that a component library containing various implementations of elementary circuit components is available. An open problem is how to construct such a component library in the context of approximate computing, where the component's error is a new design variable and hence many compromise implementations exist for a given component. In this paper, we first introduce a multi-objective Cartesian genetic programming method to create a comprehensive component library containing hundreds of Pareto optimal implementations of approximate 8-bit adders and multipliers, where the error, area and delay are simultaneously optimized. Another multi-objective evolutionary algorithm is employed to solve the so called binding problem of HLS, in which suitable approximate components are assigned to nodes of the data flow graph describing a complex digital circuit. Two approaches are then proposed and compared in order to reduce the size of the library of approximate components.  It is shown that a random subsampling of the component library provides satisfactory results in the context of our study. The proposed methods are evaluated using two benchmark circuits -- the reduce (sum) and DCT circuits.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20206 - Computer hardware and architecture

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2016

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2016 IEEE Symposium Series on Computational Intelligence

ISBN

978-1-5090-4240-1

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

1-8

Název nakladatele

IEEE Computational Intelligence Society

Místo vydání

Athens

Místo konání akce

Athens

Datum konání akce

6. 12. 2016

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000400488302074