A parallel similarity search in high dimensional metric space using M-tree

ALPKOÇAK A., Danisman T., Ulker T.

NATO Advanced Research Workshop on Advanced Environments, Tools and Applications for Cluster Computing, IWCC 2001, Mangalia, Romania, 1 - 06 September 2001, vol.2326, pp.166-171 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 2326
  • Doi Number: 10.1007/3-540-47840-x_16
  • City: Mangalia
  • Country: Romania
  • Page Numbers: pp.166-171
  • Dokuz Eylül University Affiliated: Yes

Abstract

© Springer-Verlag Berlin Heidelberg 2002.In this study, parallel implementation of M-tree to index high dimensional metric space has been elaborated and an optimal declustering technique has been proposed. First, we have defined the optimal declustering and developed an algorithm based on this definition. Proposed declustering algorithm considers both object proximity and data load on disk/processors by executing a k-NN or a range query for each newly inserted objects. We have tested our algorithm in a database containing randomly chosen 1000 image’s color histograms with 32 bins in HSV color space. Experimentation showed that our algorithm produces a very near optimal declustering.