Storage And Processing Of Massive Data In The Power Distribution Network Based On Hadoop

With the increase of power system scale and complexity,the installation quantity of monitoring terminal in the power distribution network is increasing,the data collected are therefore linear rising.In the current,processing system of power distribution network data is mainly using relational database and the centralized treatment which appear to be low in terms of computation efficiency facing large data,making the great data quality poor.Therefore,solving the problem of mass power distribution network data storage and improving the computational efficiency of huge amounts of power data is particularly important in the development of electric power system in China.While current grid of traditional database couldn,t not satisfy the storage of mass data and the calculation efficiency was low,to solve this problem,this paper proposed a storage and processing solution to the mass index data of power distribution network based on Hadoop by studying MapReduce(the parallel computing framework)and HDFS(Hadoop Distributed FileSystem)which were in the platform of Hadoop,and the HBase(distributed database based on the column),meanwhile referring to present power grid data storage method.