MR-tree: An efficient index for MapReduce

Abstract

Nowadays, big data becomes more and more popular, because it widely exists in many applications, such as social network and astronomy. Although building indexes to improve the query processing performance is common in DBMS field, it is infeasible to apply traditional indexing techniques to the MapReduce framework efficiently. Thus, how to process such data efficiently is challenging. In this paper, we study the problem of how to build an index for multidimensional data in MapReduce platform. The experimental results show that the proposed method can run efficiently. Copyright © 2013 John Wiley & Sons, Ltd.