A Distributed Storage Strategy For Trajectory Data Based On Nosql Database
Yan Jie Zhou, Qifan Chen, Baoyu Shan, Fan Jiang, Yuling Pang · 2019
The trajectory data often contains a large amount of information. So effective storage management is needed to analyze and apply the trajectory data. This paper proposes a distributed storage strategy for the characteristics of trajectory data and the requirement of data balance among distributed nodes. A trajectory data is converted into a rectangle by the MBR to ensure data integrity; and the center of the MBR is calculated for K-Means clustering; then, this paper divides the clustering results by grid and merges the grids with high overlap rate. Finally, a data partitioning strategy is used to store the divided data in each nodes so that the data volume of each node is balance. Experiments show that the proposed distributed storage strategy can effectively take into account the integrity of trajectory data, spatial proximity, data structure flexibility to achieve data balance among each nodes based on distributed NoSQL database.