Range Queries in Spatial Index Research Based on the Spark

Big data processing system has a good ability of data storage and data processing, can satisfy the demand of the processing of large amounts of data, efficient and high performance processing large data sets, the value of the data. In order to perform more efficient high-performance data retrieval, proposed to Spark system optimization, based on the Spark system, extend Spark SQL components and absorbing the thought of SIMBA, introduced two new global filtering and local spatial index optimization strategy, can make the system with high throughput and low delay to execute the temporal query operation. Evaluation experiments showed that the optimized system has the ability of high performance index of process data effectively and efficiently.