StarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

StarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators

Abstract

GPUs clusters are becoming widespread HPC platforms. Ex- ploiting them is however challenging, as this requires two separate paradigms (MPI and CUDA or OpenCL) and careful load balancing due to node heterogeneity. Current paradigms usually either limit themselves to of- fload part of the computation and leave CPUs idle, or require static CPU/GPU work partitioning. We thus have previously proposed StarPU, a runtime system able to dynamically scheduling tasks within a single heterogeneous node. We show how we extended the task paradigm of StarPU with MPI to easily map the task graph on MPI clusters and automatically benefit from optimized execution.
Fichier principal
Vignette du fichier
starpu_mpi_poster.pdf (154.11 Ko) Télécharger le fichier
poster_eurompi_2012.pdf (11.74 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-00725477 , version 1 (16-08-2021)

Identifiers

  • HAL Id : hal-00725477 , version 1

Cite

Cédric Augonnet, Olivier Aumage, Nathalie Furmento, Raymond Namyst, Samuel Thibault. StarPU-MPI: Task Programming over Clusters of Machines Enhanced with Accelerators. EuroMPI 2012 - The 19th European MPI Users' Group Meeting, Sep 2012, Vienna, Austria. ⟨hal-00725477⟩
556 View
584 Download

Share

Gmail Facebook Twitter LinkedIn More