Predicting number of threads using balanced datasets for openMP regions
Alcaraz, Jordi (Universitat Autònoma de Barcelona. Departament d'Arquitectura de Computadors i Sistemes Operatius)
TehraniJamsaz, Ali (Iowa State University)
Dutta, Akash (Iowa State University)
Sikora, Anna (Universitat Autònoma de Barcelona. Departament d'Arquitectura de Computadors i Sistemes Operatius)
Jannesari, Ali (Iowa State University)
Sorribes Gomis, Joan (Universitat Autònoma de Barcelona. Departament d'Arquitectura de Computadors i Sistemes Operatius)
César Galobardes, Eduardo (Universitat Autònoma de Barcelona. Departament d'Arquitectura de Computadors i Sistemes Operatius)

Fecha:	2023
Resumen:	Incorporating machine learning into automatic performance analysis and tuning tools is a promising path to tackle the increasing heterogeneity of current HPC applications. However, this introduces the need for generating balanced datasets of parallel applications' executions and for dealing with natural imbalances for optimizing performance parameters. This work proposes a holistic approach that integrates a methodology for building balanced datasets of OpenMP code-region patterns and a way to use such datasets for tuning performance parameters. The methodology uses hardware performance counters to characterize the execution of a given region and correlation analysis to determine whether it covers an unique part of the pattern input space. Nevertheless, a balanced dataset of region patterns may become naturally imbalanced when used for training a model for tuning any specific performance parameter. For this reason, we have explored several methods for dealing with naturally imbalanced datasets for finding the appropriated way of using them for tuning purposes. Experimentation shows that the proposed methodology can be used to build balanced datasets and that such datasets, plus a combination of Random Forest and binary classification, can be used to train a model able to accurately tune the number of threads of OpenMP parallel regions.
Ayudas:	Agencia Estatal de Investigación PID2020-113614RB-C21 Agència de Gestió d'Ajuts Universitaris i de Recerca 2017/SGR-313
Nota:	Altres ajuts: acords transformatius de la UAB
Derechos:	Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, fins i tot amb finalitats comercials, sempre i quan es reconegui l'autoria de l'obra original.
Lengua:	Anglès
Documento:	Article ; Versió publicada
Materia:	Hardware performance counters ; Machine learning ; Parallel applications ; Performance tuning ; OpenMP
Publicado en:	Computing, Vol. 105 (May 2023) , p. 999-1017, ISSN 1436-5057

DOI: 10.1007/s00607-022-01081-6

19 p, 1.4 MB

El registro aparece en las colecciones:
Artículos > Artículos publicados