Research On Cloud Storage Technology Oriented To Measurement Data Of Smart Grid

With the continuous development of smart grid and update of measurement equipment, the size of the measurement data has soared from the original GB grade level to the current PB grade level. It requests that these massive data must be stored securely and effectively. Cloud storage technology can be used to store vast amounts of data for the smart grid.This paper presents a cloud storage system oriented to measurement data of the smart grid and completes the following work:(1)This paper analyzes the characteristics of the measurement data of smart grid and improves the original cloud storage system model. Object-based storage technology has been added at the bottom. Object-based storage has many features, for example versatility, high performance, high reliability and sharing features. Client can directly access to the object-based storage server. Related metadata information only needs to access a file when you first opened. When you firstly access to the file, related metadata only need to access once. Thereby it eliminates bottlenecks of logic-based storage.(2) The measurement data of the smart grid has the characteristics of real time and great magnanimity. It requires large-capacity and high-performance large-scale storage system to store the massive data. However, the large-scale storage system is developed step by step and is formed by thousands of torage devices. There is a big difference in the performance between the new device and the old device. Therefore, the data placement algorithm is one of the key issues, efficiently organizing storage devices. This paper based on the object-based storage system model introduces the hierarchical data placement algorithm and proposes new device weight setting mode and clustering method. It reduces the weight difference between heterogeneous devices through clustering.(3) Individual measurement system of smart grid requires that the data is stored securely. Data integrity is a basic requirement of data storage security. This paper based on the object storage system model improves the existing PDP scheme. The initialization phase is moved to the cloud. It uses BLS short signature for data integrity verification, and introduces Merkle hash tree to quickly find destroy data And finally it achieves the goal that secure storage of individual measurement data.(4) This paper analyzes the hierarchical data placement algorithm proposed in this paper through experiment. Firstly, the experiment verify its fairness and adaptability of the algorithm. Secondly, the experiment tests the time effectiveness and space effectiveness of the algorithm.