Parallel asynchronous modelization and execution of cholesky algorithm using petri nets

Parallelization of algorithms with hard data dependency has a need of task synchronization. Synchronous parallel versions are simple to model and program, but inefficient in terms of scalability and processors use rate. The same problem for Asynchronous versions with elemental static task scheduling...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Wolfmann, Gustavo, De Giusti, Armando
Formato: conferenceObject
Lenguaje:Inglés
Publicado: 2022
Materias:
Asynchronous Parallel Execution
Dynamic Task Scheduling
GPGPU’s
Acceso en línea:http://hdl.handle.net/11086/28080
Aporte de:
Repositorio Digital Universitario (UNC) de Universidad Nacional de Córdoba
id I10-R141-11086-28080
record_format dspace
institution Universidad Nacional de Córdoba
institution_str I-10
repository_str R-141
collection Repositorio Digital Universitario (UNC)
language Inglés
topic Asynchronous Parallel Execution
Dynamic Task Scheduling
GPGPU’s
spellingShingle Asynchronous Parallel Execution
Dynamic Task Scheduling
GPGPU’s
Wolfmann, Gustavo
De Giusti, Armando
Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
topic_facet Asynchronous Parallel Execution
Dynamic Task Scheduling
GPGPU’s
description Parallelization of algorithms with hard data dependency has a need of task synchronization. Synchronous parallel versions are simple to model and program, but inefficient in terms of scalability and processors use rate. The same problem for Asynchronous versions with elemental static task scheduling. Efficient Asynchronous algorithms implement out-of-order execution and are complex to model and execute. In this paper we introduce Petri Nets as a tool for simplifying the modeling and execution of parallel asynchronous versions of this kind of algorithms, while using an efficient dynamic task scheduling implementation. The Cholesky factorization algorithm was used as testbed. Simulations were carried out as a proof of concept, based on real execution times on GPGPU’s, and haveshown excellent performances.
format conferenceObject
author Wolfmann, Gustavo
De Giusti, Armando
author_facet Wolfmann, Gustavo
De Giusti, Armando
author_sort Wolfmann, Gustavo
title Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
title_short Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
title_full Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
title_fullStr Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
title_full_unstemmed Parallel asynchronous modelization and execution of cholesky algorithm using petri nets
title_sort parallel asynchronous modelization and execution of cholesky algorithm using petri nets
publishDate 2022
url http://hdl.handle.net/11086/28080
work_keys_str_mv AT wolfmanngustavo parallelasynchronousmodelizationandexecutionofcholeskyalgorithmusingpetrinets
AT degiustiarmando parallelasynchronousmodelizationandexecutionofcholeskyalgorithmusingpetrinets
bdutipo_str Repositorios
_version_ 1764820391838089221

Ejemplares similares