Research Of Spatial Data Cloud Storage Based On NoSQL Database

GIS is a system for the data acquisition, storage, computing, describe and display which had beenabstract under the support of computer technology. It is widely applied in the fields of transportation, urbanplanning, commercial finance, resource management. The spatial data has the following characteristics,Spatial data is a data describe geospatial entities exist in reality with points, lines, surfaces. The amount ofthe spatial data is very large. The data needs comprehensive, classified to described abstractly the spatialentity before storing. Generally, once the data is formed is not easily modified, If you want to modify,onlythe batch modify. Storing data having a certain period of time. GIS constantly changing, the data needs tobe updated. On the basis of a relational database, the traditional spatial data storage systems through spatialdata model extensions to store massive amounts of spatial data, spatial data and attribute data are storedseparately. This storage mode exists poor scalability, low processing efficiency, lack of performance dataaccess limitations.The characteristics of spatial data and the limitations of traditional relational database to store spatialdata prompted us to find a better storage strategy to store the data, The rise of geographic informationtechnology, led to the development of geospatial data storage technology. Currently, the widely used of web2.0technologies, geospatial data cloud storage is also flourish with an imperative gesture. Therefore, Thespatial data in an advanced storage strategy efficiently stored in the cloud, is bound to bring the whole paceof geographic information systems. Related industries will also become the research hotspot. On this basis,based on a distributed NoSQL database storage space under the cloud computing environment dataprogram has been widely studied and applied. Thus, NoSQL distributed database solution data storagespatial data can become wide range of research and application in a cloud environment.According to the research, this paper do the work as follows:(1) MongDB have document-oriented storage, full index support, high availability, auto-sharding,weak consistency (final agreement) and other characteristics, these features can improve the reliability andefficiency of access to access vast amounts of data. Based on this point, the use of MongoDB database tostore massive amounts of spatial data, To ensure that the spatial data high availability, scalability, security and flexibility, do not need the complex extended operation like the relational database to store large datasimplified the complexity of traditional relational data access operations.(2) Analysis of the different between distributed database MongoDB and Traditional relationaldatabase. Made excellent points and innovative of the stores for geospatial data. Elaborate the technical,theoretical anchor for why to choose MongoDB.(3) Use MongoDB distributed deployment scenarios, spatial data will be deployed to the cluster serveras shards. Generation spatial data by abstract the geographic entity. On the basis of the K-Means algorithmto classify the generated data then stored in a different shards. Achieve the system that to read,write andInquiry the geospatial data that stored in the cloud to verify the performance of the operation and efficiencyof data for this storage solution, then comprehensive analysis of the data, draw relevant conclusions.through the Interface System to verify the MongoDB cloud-based storage solution is efficient and feasible.