skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Parallel k-means++ for Multiple Shared-Memory Architectures

Abstract

In recent years k-means++ has become a popular initialization technique for improved k-means clustering. To date, most of the work done to improve its performance has involved parallelizing algorithms that are only approximations of k-means++. In this paper we present a parallelization of the exact k-means++ algorithm, with a proof of its correctness. We develop implementations for three distinct shared-memory architectures: multicore CPU, high performance GPU, and the massively multithreaded Cray XMT platform. We demonstrate the scalability of the algorithm on each platform. In addition we present a visual approach for showing which platform performed k-means++ the fastest for varying data sizes.

Authors:
;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1334876
Report Number(s):
PNNL-SA-116273
DOE Contract Number:
AC05-76RL01830
Resource Type:
Conference
Resource Relation:
Conference: 45th International Conference on Parallel Processing (ICPP 2016), August 16-19, 2016, Philadelphia, PA, 93-102
Country of Publication:
United States
Language:
English
Subject:
data clustering; GPGPU; multithreaded architecture; OpenMP

Citation Formats

Mackey, Patrick S., and Lewis, Robert R.. Parallel k-means++ for Multiple Shared-Memory Architectures. United States: N. p., 2016. Web. doi:10.1109/ICPP.2016.18.
Mackey, Patrick S., & Lewis, Robert R.. Parallel k-means++ for Multiple Shared-Memory Architectures. United States. doi:10.1109/ICPP.2016.18.
Mackey, Patrick S., and Lewis, Robert R.. Thu . "Parallel k-means++ for Multiple Shared-Memory Architectures". United States. doi:10.1109/ICPP.2016.18.
@article{osti_1334876,
title = {Parallel k-means++ for Multiple Shared-Memory Architectures},
author = {Mackey, Patrick S. and Lewis, Robert R.},
abstractNote = {In recent years k-means++ has become a popular initialization technique for improved k-means clustering. To date, most of the work done to improve its performance has involved parallelizing algorithms that are only approximations of k-means++. In this paper we present a parallelization of the exact k-means++ algorithm, with a proof of its correctness. We develop implementations for three distinct shared-memory architectures: multicore CPU, high performance GPU, and the massively multithreaded Cray XMT platform. We demonstrate the scalability of the algorithm on each platform. In addition we present a visual approach for showing which platform performed k-means++ the fastest for varying data sizes.},
doi = {10.1109/ICPP.2016.18},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu Sep 22 00:00:00 EDT 2016},
month = {Thu Sep 22 00:00:00 EDT 2016}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share: