Multi-Dimensional Indexing System Considering Distributions of Sub-indexes and Data
- https://doi.org/10.2991/ijndc.k.191118.002How to use a DOI?
- Multi-dimensional index, parallel processing, retrieval performance, dissimilarity, distribution
This paper experimentally evaluates the parallel multi-dimensional indexing system indexing data by using several multi-dimensional indexes and retrieving required data from them in parallel. A constant number of data are indexed into a sub-index. The sub-index is inserted into an index in a computational node. After the number of sub-indexes in a computational node is evaluated, the area managed by each index is evaluated so that the numbers of sub-indexes become equal, and the data are widely distributed. It is experimentally shown that this method has good performance for skewed data.
- © 2019 The Authors. Published by Atlantis Press SARL.
- Open Access
- This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - JOUR AU - Kazuto Nakanishi AU - Teruhisa Hochin AU - Hiroki Nomiya AU - Hiroaki Hirata PY - 2019 DA - 2019/12 TI - Multi-Dimensional Indexing System Considering Distributions of Sub-indexes and Data JO - International Journal of Networked and Distributed Computing SP - 25 EP - 33 VL - 8 IS - 1 SN - 2211-7946 UR - https://doi.org/10.2991/ijndc.k.191118.002 DO - https://doi.org/10.2991/ijndc.k.191118.002 ID - Nakanishi2019 ER -