Type-driven automated program transformations and cost modelling for optimising streaming programs on FPGAs

Vanderbauwhede, W. , Nabi, S. W. and Urlea, C. (2019) Type-driven automated program transformations and cost modelling for optimising streaming programs on FPGAs. International Journal of Parallel Programming, 47(1), pp. 114-136. (doi: 10.1007/s10766-018-0572-z)

[img]
Preview
Text
148817.pdf - Published Version
Available under License Creative Commons Attribution.

1MB

Abstract

In this paper we present a novel approach to program optimisation based on compiler-based type-driven program transformations and a fast and accurate cost/performance model for the target architecture. We target streaming programs for the problem domain of scientific computing, such as numerical weather prediction. We present our theoretical framework for type-driven program transformation, our target high-level language and intermediate representation languages and the cost model and demonstrate the effectiveness of our approach by comparison with a commercial toolchain.

Item Type:Articles
Additional Information:The authors acknowledge the support of the UK EPSRC for the TyTra project (EP/L00058X/1).
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Urlea, Mr Cristian and Vanderbauwhede, Professor Wim and Nabi, Dr Syed Waqar
Authors: Vanderbauwhede, W., Nabi, S. W., and Urlea, C.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:International Journal of Parallel Programming
Publisher:Springer
ISSN:0885-7458
ISSN (Online):1573-7640
Published Online:25 April 2018
Copyright Holders:Copyright © 2018 The Authors
First Published:First published in International Journal of Parallel Programming 47:114–136
Publisher Policy:Reproduced under a Creative Commons License

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
614451Exploiting Parallelism through Type Transformations for Hybrid Manycore Systems.Wim VanderbauwhedeEngineering and Physical Sciences Research Council (EPSRC)EP/L00058X/1COM - COMPUTING SCIENCE