Detailed Analysis and Optimization of CUDA K-means Algorithm

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F20%3A10414260" target="_blank" >RIV/00216208:11320/20:10414260 - isvavai.cz</a>

Výsledek na webu

<a href="https://dl.acm.org/doi/abs/10.1145/3404397.3404426" target="_blank" >https://dl.acm.org/doi/abs/10.1145/3404397.3404426</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3404397.3404426" target="_blank" >10.1145/3404397.3404426</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Detailed Analysis and Optimization of CUDA K-means Algorithm

Popis výsledku v původním jazyce

K-means is one of the most frequently used algorithms for unsupervised clustering data analysis. Individual steps of the k-means algorithm include nearest neighbor finding, efficient distance computation, and cluster-wise reduction, which may be generalized to many other purposes in data analysis, visualization, and machine learning. Efficiency of the available implementations of k-means computation steps therefore directly affect many other applications. In this work, we examine the performance limits in the context of modern massively parallel GPU accelerators. Despite the existence of many published papers on this topic, we have found that crucial performance aspects of the GPU implementations remain unaddressed, including the optimizations for memory bandwidth, cache limits, and workload dispatching on problem instances of varying cluster count, dataset size, and dimensionality. We present a detailed analysis of individual computation steps and propose several optimizations that improve the overall performance on contemporary GPU architectures. Our open-source prototype exhibits significant speedup over the current state-of-the-art implementations in virtually all practical scenarios.

Název v anglickém jazyce

Detailed Analysis and Optimization of CUDA K-means Algorithm

Popis výsledku anglicky

K-means is one of the most frequently used algorithms for unsupervised clustering data analysis. Individual steps of the k-means algorithm include nearest neighbor finding, efficient distance computation, and cluster-wise reduction, which may be generalized to many other purposes in data analysis, visualization, and machine learning. Efficiency of the available implementations of k-means computation steps therefore directly affect many other applications. In this work, we examine the performance limits in the context of modern massively parallel GPU accelerators. Despite the existence of many published papers on this topic, we have found that crucial performance aspects of the GPU implementations remain unaddressed, including the optimizations for memory bandwidth, cache limits, and workload dispatching on problem instances of varying cluster count, dataset size, and dimensionality. We present a detailed analysis of individual computation steps and propose several optimizations that improve the overall performance on contemporary GPU architectures. Our open-source prototype exhibits significant speedup over the current state-of-the-art implementations in virtually all practical scenarios.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

ICPP &apos;20: Proceedings of the 49th International Conference on Parallel Processing

ISBN

978-1-4503-8816-0

ISSN

—

e-ISSN

—

Počet stran výsledku

11

Strana od-do

—

Název nakladatele

ACM

Místo vydání

New York, NY, USA

Místo konání akce

Edmonton AB Canada

Datum konání akce

17. 8. 2020

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—