Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions

Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions

We present GraSSP, a novel approach to perform automated parallelization relying on recent advances in formal verification and synthesis. GraSSP augments an existing sequential program with an additional functionality to decompose data dependencies in loop iterations, to compute partial results, and...

Full description

Bibliographic Details
Main Authors: Grigory Fedyukovich, Rastislav Bodík
Format: Article
Language:English
Published: Open Publishing Association 2016-11-01
Series:Electronic Proceedings in Theoretical Computer Science
Online Access:http://arxiv.org/pdf/1611.07629v1
id doaj-40666981e6ce4773b2ba4f2bd5e2ab0f
record_format Article
spelling doaj-40666981e6ce4773b2ba4f2bd5e2ab0f2020-11-24T22:35:53ZengOpen Publishing AssociationElectronic Proceedings in Theoretical Computer Science2075-21802016-11-01229Proc. SYNT 2016556610.4204/EPTCS.229.6:10Approaching Symbolic Parallelization by Synthesis of Recurrence DecompositionsGrigory Fedyukovich0Rastislav Bodík1 UW UW We present GraSSP, a novel approach to perform automated parallelization relying on recent advances in formal verification and synthesis. GraSSP augments an existing sequential program with an additional functionality to decompose data dependencies in loop iterations, to compute partial results, and to compose them together. We show that for some classes of the sequential prefix sum problems, such parallelization can be performed efficiently.http://arxiv.org/pdf/1611.07629v1
collection DOAJ
language English
format Article
sources DOAJ
author Grigory Fedyukovich
Rastislav Bodík
spellingShingle Grigory Fedyukovich
Rastislav Bodík
Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
Electronic Proceedings in Theoretical Computer Science
author_facet Grigory Fedyukovich
Rastislav Bodík
author_sort Grigory Fedyukovich
title Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
title_short Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
title_full Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
title_fullStr Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
title_full_unstemmed Approaching Symbolic Parallelization by Synthesis of Recurrence Decompositions
title_sort approaching symbolic parallelization by synthesis of recurrence decompositions
publisher Open Publishing Association
series Electronic Proceedings in Theoretical Computer Science
issn 2075-2180
publishDate 2016-11-01
description We present GraSSP, a novel approach to perform automated parallelization relying on recent advances in formal verification and synthesis. GraSSP augments an existing sequential program with an additional functionality to decompose data dependencies in loop iterations, to compute partial results, and to compose them together. We show that for some classes of the sequential prefix sum problems, such parallelization can be performed efficiently.
url http://arxiv.org/pdf/1611.07629v1
work_keys_str_mv AT grigoryfedyukovich approachingsymbolicparallelizationbysynthesisofrecurrencedecompositions
AT rastislavbodik approachingsymbolicparallelizationbysynthesisofrecurrencedecompositions
_version_ 1725722588199518208

Similar Items