Research And Implementation Of Massive Data Disaster Recovery Based On Data Summarisation Technology

With the rapid development of information technology, today's society has entered into a data explosion era of massive data. Business system data scale is growing. On the other hand, with the rapid enhancement of data mining and utilization capacity, the value of the data is also rising. Therefore, the data security protection becomes more and more important, how to avoid disaster, or how quickly after a disaster recovery service capabilities, reduce operational losses becomes academia, industry, and government departments at all levels to focus and research issue.The Disaster Tolerant(DT) is an important means of protection and technology for information systems. Disaster Recovery refers to the process of business systems to ensure the normal operation of business continuity during a disaster accident. The core purpose is to prevent business systems disaster recovery in the event of a disaster cause system outages and data loss. Disaster mostly outside the main system through the establishment and maintenance of a backup system, the use of geographical isolation to ensure security of business systems. Traditional disaster recovery methods are through the entirely data backup and recovery to protect data security and reliability. However, traditional disaster recovery technology in the face of massive data, since the data disaster recovery time is too long, seriously affecting the availability of business systems, often resulting in business a long break and a huge loss.To solve these problems, this paper analyzes the basis of traditional disaster recovery technology and data summarization on technology-related technology to rapidly operational capacity massive database system disaster recovery after the goal, studied a data summarization technology for disaster recovery, design and implementation of a distributed data management platform for disaster recovery. Details are as follows:1. We design a rapid recovery system core business empowerment programs. First recover business-critical data and summary data for fast recovery core business competencies, and then restore other data.2. We propose a method for fast data summarization for disaster recovery. Proposed a two-stage compression of data summaries technology, greatly reducing the size of backup data, while retaining the characteristic pattern information of the original data.3. We design and implement a distributed data management platform for disaster recovery. Benefit of its core functionality to build a data summary data subsystem data summarization for efficient data disaster recovery, while the backup data with high reliability and high efficiency of storage management, enabling efficient disaster recovery.