2026-06-05T14:23:15Zhttps://riuma.uma.es/rest/oai/request

oai:riuma.uma.es:10630/138912026-02-03T12:09:57Zcom_10630_2254col_10630_37959

Canales, Marcos Cáncer, Jorge Constantinescu, Denisa-Andreea Escuin, Carlos Perez, Borja 2017-06-15T10:12:02Z 2017-06-15T10:12:02Z 2017-06-15 http://hdl.handle.net/10630/13891 Clustering is the task of assigning a set of objects into groups (clusters) so that objects in the same group are more similar to each other than to those in other groups. In particular, K-means is a clustering algorithm that calculates the cluster with the nearest mean for each object. To achieve this, it uses a function like Euclidean or Manhattan distance. Our objective is to exploit our heterogeneous computing environment, that integrates an Intel Core i7-6700K chip, 2x NVIDIA TITAN X and an Intel Altera Terasic Stratix V DE5-NET FPGA, to run K-means as fast as possible. eng open access by-nc-nd Algoritmos computacionales Three is not a crowd: ACPU-GPU-FPGA K-means implementation conference output