Application Research Of Cloud Computing And Data Disaster Recovery In Electric Power System

In recent years, with the rapid development of information technology, the data has became an essential part for enterprises. As an important role for people?s living hood, data for electric power enterprise is extremely important and will leave a great loss once it?s damaged. How to prevent the electric power industry data get damaged from fire, earthquake, war and other disasters, and on the other hand, make sure the continuity and security business, which has became an important question to be solved.In recent years, electric power enterprises have established an unified set of disaster recovery center, so as to enhance the security and stability of the whole power grid. However, with the rapid development of information technology, especially the coming of ?Big Data Time? and ?Cloud Computing Time?, traditional data disaster recovery scheme can hardly catch up with times because of its poor scalability, low utilization rate of resources and so on. What?s more, the electric power data has sharply increased with each passing day, and it has reached TB even PB level for daily incremental. That has lead a large increase in cost for traditional data disaster recovery. In addition, such great amount of data can also leave a serious influence for the time of data disaster recovery.This paper has analysed the existing recovery framework and key technologies firstly. According to the extension problem about disaster recovery resources of dispatching data and the long data recovery time problem. The paper has combined with cloud computing and data disaster recovery technology, using the “cloud model” based on the existing disaster recovery center of power system and combined with multi-objective particle swarm optimization algorithm in dynamic environment, it presents a new strategy of data disaster recovery in power system. By this way, the extension problem about disaster recovery resources of dispatching data is solved on one hand, and has considered both the disaster recovery cost and disaster recovery time on another. Finally, the validity of the strategy is verified by experiments.