An Efficient Architecture for Parallel Skyline Computation over Large Distributed Datasets

Abstract

The skyline query has been received much attention as an important operator in database systems for multi-preference analysis and decision making. Most of the previous works have focused on processing skyline queries on centralized data sets. However, the related data in real applications are practically scattered at several different servers. The skyline query computation in distributed environments is needed to gather a large number of data from the connected servers. The existing methods for a skyline query in distributed environments have two problems: (i) They have slow processing time for a skyline query. (ii) Most of the transferred data among servers in the network are unnecessary. In this paper, we propose a multi-layer grid method for efficiently processing a skyline query in distributed environments. The proposed method minimizes the unnecessary transferred data by using the grid mechanism. Experiments based on synthetic data sets and real data sets show that our proposed method outperforms the existing methods.