Research On Access Optimization Method Of Spatiotemporal Big Data

In an era where everything can be connected,various devices have been equipped with the functionality to record data that contains time and space,as well as all kinds of attribute states,which can be collectively referred to as spatio-temporal data.How to make the most of this data is the current challenge.The first problem is how to access this data.Traditional spatio-temporal databases face the problem of not being able to undertake such large amounts of data operations and unstructured data.The current big data access solutions can solve problems such as large data volume and unstructured data,but they do not support spatio-temporal operation.Therefore,there is a need to urgently solve the problem of spatio-temporal big data access for analyzing and processing spatio-temporal big data.The current spatiotemporal big data access are mainly solved from two aspects,that is,by closely combining with the big data system framework and designing the big data access table.However,these two solutions can only solve parts of the problem,and the access efficiency cannot be satisfactory.After analyzing the data access solution of big data and the traditional spatiotemporal data access solution,this paper devised an optimization based on the existing spatio-temporal big data access solution.A data access solution that combines Geohashes access index based on XZ-order with the optimized HBase table was also proposed.The main research contents are as follows:1.The index solution of the traditional relational data for spatio-temporal data access was deeply analyzed.In addition,the characteristics and advantages of the big data access plan were analyzed.2.According to the current spatio-temporal data access solution,a kind of spatio-temporal data index based on the XZ-order was proposed and an HBase table structure suitable for spatio-temporal data access was designed.Through the correspondence between the spatio-temporal dimensions and the key in the list structure in the index,the query efficiency of spatio-temporal data in the spatiotemporal dimension would be improved.Furthermore,environmental construction and comprehensive testing that target the proposed solution were performed.3.Comparative experiments on the optimization solution were taken by selecting point,line,and surface geographic entity data,so as to test the solution proposed in this paper.