Please use this identifier to cite or link to this item: http://hdl.handle.net/10995/35997
Title: Partition Around Medoids Clustering on the Intel Xeon Phi Many-Core Coprocessor
Authors: Rechkalov, T. V.
Issue Date: 2015
Publisher: Уральский федеральный университет
Citation: Rechkalov T. V. Partition Around Medoids Clustering on the Intel Xeon Phi Many-Core Coprocessor / T. V. Rechkalov // Proceedings of the 1st Ural Workshop on Parallel, Distributed, and Cloud Computing for Young Scientists (Ural-PDC 2015). — Yekaterinburg, 2015. — P. 29-41. — (CEUR Workshop Proceedings ; vol. 1513).
Abstract: Abstract. The paper touches upon the problem of implementation Partition Around Medoids (PAM) clustering algorithm for the Intel Many Integrated Core architecture. PAM is a form of well-known k-Medoids clustering algorithm and is applied in various subject domains, e.g. bioinformatics, text analysis, intelligent transportation systems, etc. An optimized version of PAM for the Intel Xeon Phi coprocessor is introduced where OpenMP parallelizing technology, loop vectorization, tiling technique and efficient distance matrix computation for Euclidean metric are used. Experimental results for different data sets confirm the efficiency of the proposed algorithm.
Keywords: DATA MINING
CLUSTERING
K-MEDOIDS
PARTITION AROUND MEDOIDS
INTEL MANY INTEGRATED CORE ARCHITECTURE
INTEL XEON PHI COPROCESSOR
PARALLEL COMPUTING
TILING
VECTORIZATION
OPENMP
URI: http://hdl.handle.net/10995/35997
Conference name: 1st Ural Workshop on Parallel, Distributed, and Cloud Computing for Young Scientists (Ural-PDC 2015)
Conference date: 17.11.2015
ISSN: 1613-0073
Origin: CEUR Workshop Proceedings. Vol. 1513 : Proceedings of the 1st Ural Workshop on Parallel, Distributed, and Cloud Computing for Young Scientists (Ural-PDC 2015). — Yekaterinburg, 2015
Appears in Collections:Конференции, семинары

Files in This Item:
File Description SizeFormat 
ural_pdc-2015-04.pdf868,86 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.