Parallelization of the k-means Algorithm in a Spectral Clustering Chain on CPU-GPU Platforms - Laboratoire de recherche en informatique. Équipe: Systèmes Parallèles
Conference Papers Year : 2020

Parallelization of the k-means Algorithm in a Spectral Clustering Chain on CPU-GPU Platforms

Parallélisation d'un algorithme de k-means dans une suite logicielle de Spectral Clustering sur plateforme CPU+GPU

Abstract

k-means is a standard algorithm for clustering data. It constitutes generally the final step in a more complex chain of high quality spectral clustering. However this chain suffers from lack of scalability when addressing large datasets. This can be overcome by applying also the k-means algorithm as a pre-processing task to reduce the input data instances. We describe parallel optimization techniques for the k-means algorithm on CPU and GPU. Experimental results on synthetic dataset illustrate the numerical accuracy and performance of our implementations .
Fichier principal
Vignette du fichier
HeteroPar2020_He-Vialle-Baboulin.pdf (456.41 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02985021 , version 1 (01-11-2020)

Identifiers

  • HAL Id : hal-02985021 , version 1

Cite

Guanlin He, Stéphane Vialle, Marc Baboulin. Parallelization of the k-means Algorithm in a Spectral Clustering Chain on CPU-GPU Platforms. HeteroPar Workshop of 2020 Euro-Par International Conference, Aug 2020, Warsaw, Poland. ⟨hal-02985021⟩
163 View
275 Download

Share

More