Fine-grained MPI+OpenMP plasma simulations: communication overlap with dependent tasks - Maison de la Simulation Accéder directement au contenu
Article Dans Une Revue Lecture Notes in Computer Science Année : 2019

Fine-grained MPI+OpenMP plasma simulations: communication overlap with dependent tasks

Résumé

This paper demonstrates how OpenMP 4.5 tasks can be used to efficiently overlap computations and MPI communications based on a case-study conducted on multi-core and many-core architectures. It focuses on task granularity, dependencies and priorities, and also identifies some limitations of OpenMP. Results on 64 Skylake nodes show that while 64% of the wall-clock time is spent in MPI communications, 60% of the cores are busy in computations, which is a good result. Indeed, the chosen dataset is small enough to be a challenging case in terms of overlap and thus useful to assess worst-case scenarios in future simulations. Two key features were identified: by using task priority we improved the performance by 5.7% (mainly due to an improved overlap), and with recursive tasks we shortened the execution time by 9.7%. We also illustrate the need to have access to tools for task tracing and task visualization. These tools allowed a fine understanding and a performance increase for this task-based OpenMP+MPI code.
Fichier principal
Vignette du fichier
Richard_2019_MPI_OpenMP_communication_overlap.pdf (653.57 Ko) Télécharger le fichier
Origine : Publication financée par une institution

Dates et versions

cea-02404825 , version 1 (09-01-2023)

Identifiants

Citer

Jérôme Richard, Guillaume Latu, Julien Bigot, Thierry Gautier. Fine-grained MPI+OpenMP plasma simulations: communication overlap with dependent tasks. Lecture Notes in Computer Science, 2019, Euro-Par 2019: Euro-Par 2019: Parallel Processing, 11725, pp.419-433. ⟨10.1007/978-3-030-29400-7_30⟩. ⟨cea-02404825⟩
232 Consultations
87 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More