Study On Spatial Data Index And Query Technology Based On Phoenix Platform

Recently,with the rapid development of the social economy,numerous applications related to geographic information system such as smart bracelets,didi,and take-away meals have appeared,which results in the generation of large-scale geospatial data.People rely on spatial data in their daily lives increasingly.Therefore,the era of big spatial data has arrived.Due to the limited storage capacity of a single machine,traditional relational databases cannot deal with big data.Therefore,some distributed platforms represented by Hadoop,HBase,and Phoenix have emerged recently.Phoenix platform based on HBase has easy-to-understand two-dimensional table model,powerful SQL language,efficient secondary index,distributed horizontal extension and other excellent features,but it does not support the organization and management of spatial data.In response to the above problems,the dissertation proposes a new type of spatial data management solution named Spatial Phoenix.Spatial Phoenix selects Phoenix as the basic development platform and expands it spatially.It adds spatial operations such as spatial data preprocessing,storage,indexing,and querying.Combined with Geo Tools,it designs and implements a standard spatial data access interface,which enables the application layer to read,write,and query spatial data on Phoenix easily.The main work of the dissertation is as follows:(1)Research and design of spatial data storage model.Based on Phoenix distributed platform,the spatial data storage table model,spatial data preprocessing module and spatial data import module are designed to complete the spatial data storage operation.(2)Research and design of spatial indexing algorithm.Based on spatial data storage model,Geohash which is selected as spatial index combines with Phoenix secondary global coverage index to provide index basis for subsequent spatial queries.(3)Research and design of spatial query algorithm.Based on spatial data storage model nd spatial index algorithm,efficient spatial query algorithms are designed and implemented.They are divided into range query algorithm and k NN query algorithm,which both adapt the "three-step" strategy which includes preprocessing,filtering and refining.And both the filtering and refining are performed on the server,which can improve the spatial query efficiency greatly.(4)Design and implementation of spatial data access interface.Based on Phoenix distributed platform,spatial data access interface that interfaces with Geo Tools is designed and implemented,which enables the application layer to manage and analyze spatial data in Phoenix platform.In order to evaluate the effectiveness of Spatial Phoenix,Phoenix platform is set up.Post GIS is selected as the comparison platform.Spatial point datasets are used.Spatial data import methods and spatial query algorithms have been tested.Experimental results show that Spatial Phoenix has more efficient and more stable data import modes and more efficient spatial query efficiency when spatial datasets and query result sets reach a certain scale.