A Survey of CUDA-based Multidimensional Scaling on GPU Architecture

Authors Hasmik Osipyan, Martin Kruliš, Stéphane Marchand-Maillet



PDF
Thumbnail PDF

File

OASIcs.ICCSW.2015.37.pdf
  • Filesize: 0.56 MB
  • 9 pages

Document Identifiers

Author Details

Hasmik Osipyan
Martin Kruliš
Stéphane Marchand-Maillet

Cite AsGet BibTex

Hasmik Osipyan, Martin Kruliš, and Stéphane Marchand-Maillet. A Survey of CUDA-based Multidimensional Scaling on GPU Architecture. In 2015 Imperial College Computing Student Workshop (ICCSW 2015). Open Access Series in Informatics (OASIcs), Volume 49, pp. 37-45, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2015)
https://doi.org/10.4230/OASIcs.ICCSW.2015.37

Abstract

The need to analyze large amounts of multivariate data raises the fundamental problem of dimensionality reduction which is defined as a process of mapping data from high-dimensional space into low-dimensional. One of the most popular methods for handling this problem is multidimensional scaling. Due to the technological advances, the dimensionality of the input data as well as the amount of processed data is increasing steadily but the requirement of processing these data within a reasonable time frame still remains an open problem. Recent development in graphics hardware allows to perform generic parallel computations on powerful hardware and provides an opportunity to solve many time-constrained problems in both graphical and non-graphical domain. The purpose of this survey is to describe and analyze recent implementations of multidimensional scaling algorithms on graphics processing units and present the applicability of these algorithms on such architectures based on the experimental results which show a decrease of execution time for multi-level approaches.
Keywords
  • Dimensionality Reduction
  • Pattern Recognition
  • Graphics Hardware
  • Compute Unified Device Architecture
  • Parallel Programming

Metrics

  • Access Statistics
  • Total Accesses (updated on a weekly basis)
    0
    PDF Downloads
Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail