ALPyNA: Acceleration of Loops in Python for Novel Architectures

Jacob, D. and Singer, J. (2019) ALPyNA: Acceleration of Loops in Python for Novel Architectures. In: 6th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, Phoenix, AZ, USA, 22 Jun 2019, pp. 25-34. ISBN 9781450367172 (doi: 10.1145/3315454.3329956)

[img]
Preview
Text
188418.pdf - Accepted Version

826kB

Abstract

We present ALPyNA, an automatic loop parallelization framework for Python, which analyzes data dependences within nested loops and dynamically generates CUDA kernels for GPU execution. The ALPyNA system applies classical dependence analysis techniques to discover and exploit potential parallelism. The skeletal structure of the dependence graph is determined statically (if possible) or at runtime; this is combined with type and bounds information discovered at runtime, to auto-generate high-performance kernels for offload to GPU. We demonstrate speedups of up to 1000x relative to the native CPython interpreter across four array-intensive numerical Python benchmarks. Performance improvement is related to both iteration domain size and dependence graph complexity. Nevertheless, this approach promises to bring the benefits of manycore parallelism to application developers.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Singer, Dr Jeremy and Jacob, Dr Dejice
Authors: Jacob, D., and Singer, J.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Proceedings of the 6th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming - ARRAY 2019
Publisher:ACM Press
ISBN:9781450367172
Copyright Holders:Copyright © 2019 Association for Computing Machinery
Publisher Policy:Reproduced in accordance with the copyright policy of the publisher

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

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
614231AnyScale ApplicationsJeremy SingerEngineering and Physical Sciences Research Council (EPSRC)EP/L000725/1COM - COMPUTING SCIENCE