Research On Data Management And Analysis For Power GIS In The Cloud

Recently, GIS has become an indispensable part of electric power information construction since its unique spatial analysis capabilities and visualization skills. With the rapid development of the smart grid, the power grid expands increasingly, the electrical wiring becomes more complex, power GIS system needs to receive and process data which are wider, larger and more complicated. Traditional computing models and storage patterns are difficult to meet the needs of the power of GIS. Many limitations are gradually reflected such as GIS server insufficient computing resources and slow computing response. How to store and manage mass data efficiently is an urgent problem to be solved for power GIS.Cloud computing is a new distributed computing architecture which has good scalability and availability. It provides new solutions for power GIS. This paper applies cloud computing into powe GIS and uses Hadoop cloud platform for power GIS data storage and management. Hadoop is a well-known open source cloud computing system which provides a distributed parallel programming framework and simplifies the development of distributed applications. This paper analyzed and summarized the characteristics of various types of power GIS data, presented data storage strategy which combines relational databases and non-relational database. Designed the overall architecture of power GIS data management based on Hadoop. Designed series of data models:spatial data model which follows OGC standards, operation data model based on vertical and horizontal table structure, and other core object models. Then studied and realized the parallel data processing technologies, including tile pyramid parallel generation technology, spatial indexing parallel generation technology, parallel spatial analysis and operation data queries technology.Finally, quantities of experiments were carried out including the above parallel data processing technologies in single-machine and cluster environment to compare and validate the performance. The experimental results show that the proposed method has more obvious advantage to deal with large-scale data and has high efficiency and good feasibility. The average time of data generation, analysis and query greatly reduced after reaching certain amount of data, the application of cloud computing can meet the needs of the massive power GIS data storage and management well.