Efficient Probabilistic Skyline Query Processing in MapReduce
Linlin Ding, Guoren Wang, Junchang Xin, Ye Yuan · 2013
As a popular parallel programming model, how to process probabilistic skyline query over uncertain data in MapReduce framework is becoming an urgent problem to be resolved. In MapReduce framework, implementing probabilistic skyline query is nontrivial since the probabilistic skyline query is not decomposable. Therefore, in this paper, we propose a filter-refine two phases approach in MapReduce that translates the probabilistic skyline query into two decomposable computations for obtaining the final results. Firstly, we describe the whole processing procedure of filter-refine, and then propose an efficient probabilistic skyline query processing algorithm in MapReduce. Furthermore, to reduce the computation and communication cost, we develop the optimized probabilistic skyline query processing algorithm to prune the unpromising data both in filter and refine phases. Finally, we conduct extensive experiments on synthetic data to verify the effectiveness and efficiency of the proposed filter-refine approach with various experimental settings.