A study on MapReduce processing for multi-dimensional continuous query

Abstract

A huge volume of sensor stream data can be efficiently handle with the MapReduce framework for processing multi-dimensional continuous queries. The MapReduce originally has been used for batch processing, not real-time querying. In this paper, we propose a new idea of transforming query regions of multi-dimensional continuous queries into multiple key values. At the Map stage, key-value pairs of input data stream would be mapped into CQ-based key values that would be also grouped by the same continuous query in the Reduce stage. © Springer-Verlag 2013.