Spatial Queries Evaluation with MapReduce

Spatial queries include spatial selection query, spatial join query, nearest neighbor query, etc. Most of spatial queries are computing intensive and individual query evaluation may take minutes or even hours. Parallelization seems a good solution for such problems. However, parallel programs must communicate efficiently, balance work across all nodes, and address problems such as failed nodes. We describe MapReduce and show how spatial queries can be naturally expressed in this model, without explicitly addressing any of the details of parallelization. We present performance evaluations for several spatial queries and prove that MapReduce is also appropriate for small scale clusters and computing intensive applications.