Kernel density estimation in accelerators: implementation and performance evaluation

Lopez-Novoa, Unai, Mendiburu, Alexander and Miguel-Alonso, Jose 2016. Kernel density estimation in accelerators: implementation and performance evaluation. The Journal of Supercomputing 72 (2) , pp. 545-566. 10.1007/s11227-015-1577-7

Full text not available from this repository.

Abstract

Kernel density estimation (KDE) is a popular technique used to estimate the probability density function of a random variable. KDE is considered a fundamental data smoothing algorithm, and it is a common building block in many scientific applications. In a previous work we presented S-KDE, an efficient algorithmic approach to compute KDE that outperformed other state-of-the-art implementations, providing accurate results in much reduced execution times. Its parallel implementation targeted multi- and many-core processors. In this work we present an OpenCL implementation of S-KDE, targeting modern accelerators in a portable way. We test our implementation on three accelerators from different manufacturers, achieving speedups around 5× compared to a hand-tuned serial version of S-KDE. We also analyze the performance of the code in these accelerators, to find out to what extent our code exploits their capabilities.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Uncontrolled Keywords:	Kernel density estimation; Performance analysis; OpenCL;  Many-core processors; GPGPU
Publisher:	Springer
ISSN:	0920-8542
Last Modified:	25 Jun 2020 13:42
URI:	https://orca.cardiff.ac.uk/id/eprint/105964

Citation Data

Cited 2 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item