Optimizing Parallel System Efficiency: Dynamic Task Graph Adaptation with Recursive Tasks - Plateforme Fédérative pour la Recherche en Informatique et Mathématiques Access content directly
Conference Papers Year : 2024

Optimizing Parallel System Efficiency: Dynamic Task Graph Adaptation with Recursive Tasks

Abstract

Task-based programming models significantly improve the efficiency of parallel systems. The Sequential Task Flow (STF) model focuses on static task sizes within task graphs, but determining optimal granularity during graph submission is tedious. To overcome this, we extend StarPU’s STF recursive tasks model, enabling dynamic transformation of tasks into subgraphs. Early evaluations on homogeneous shared memory reveal that this just-in-time adaptation enhances performance.
Fichier principal
Vignette du fichier
wamta24.pdf (884.85 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04548787 , version 1 (16-04-2024)

Licence

Attribution

Identifiers

  • HAL Id : hal-04548787 , version 1

Cite

Nathalie Furmento, Abdou Guermouche, Gwenolé Lucas, Thomas Morin, Samuel Thibault, et al.. Optimizing Parallel System Efficiency: Dynamic Task Graph Adaptation with Recursive Tasks. WAMTA 2024 - Workshop on Asynchronous Many-Task Systems and Applications 2024, Feb 2024, Knoxville, United States. ⟨hal-04548787⟩
17 View
11 Download

Share

Gmail Facebook X LinkedIn More