The traditional sequential ways of spatial data processing lags in the efficiency of executing the queries. The modern parallel processing technique, the MapReduce, is used extensively for big data analysis. This paper analyzes the synthetic large spatial data-set on the MapReduce and ArcGIS to check the similarity of the outputs generated through the parallel framework and the specialized GIS software — ArcGIS. Firstly, the input data-set is processed in parallel for queries through Map and Reduce functions, and the query results are displayed on ArcGIS. Secondly, on the ArcGIS, the addresses are integrated with spatial data, through the geo-coding process that assigns addresses to locations on a map, and outputs a shapefile for display. The query results are compared for the two methods, and the similarity of the results validates our spatial data processing on the distributed compute nodes of the Hadoop cluster. The results for queries have also been compared on the scalable cluster, which shows the improved efficiency of query processing with scalability.