Research And Application Of Data Synchronization Technologies For Distributred Network

With the continuous development of computer technology and data processing capacity continues to improve, more and more complex distributed architecture is used in large information and data systems. Multi-level Distributed system compared to the conventional single centralized structure is more in line with modern organizational structure of large companies and enterprises. Distributed system helps to improve the whole system's adaptability, maintainability, modifiability and scalability. The nodes of the distributed system who have independent data processing capability locate in the different places. Systems require large amounts of data exchange and many public data duplicate deposit between them. It is very important to maintain the consistency of redundant data among the multiple nodes and data synchronization of public data. Therefore, the research of data synchronization on the distributed network is very necessary.In this paper, the data synchronization technology of distributed network is focused on. To improve the efficiency of data synchronization and improve data consistency maintenance mechanism, research on the following aspects is done.The model of peer data synchronization based on node coordination strategy is proposed. Typical of such data synchronization model has a central node responsible for distributing and managing the execution order of update transactions. This over-reliance on the fixed central node of the structure, once the central node of the structure which over-reliance on the fixed central node sent down, the whole system may collapse. To solve this problem, the model of peer data synchronization which dynamically determined the central node is proposed. The status of the nodes is completely equal. When the transaction is updated, the transaction initiated by the node is the starting point. This approach which dynamically determines the node as temporary "central node" approach is used to control the execution or rollback of the transaction. The failure of one node will only affect the submission of update transactions initiated by this node, without affecting the normal operation of the whole system. This mode improves the system ability to cope and the adaptability of the network environment and facilitates the expansion and maintenance of the system.In order to better maintain data consistency of the data synchronization model, this paper put forward solutions from three aspects. First, do research on concurrency control of data synchronization. Concurrency conflict detection of update transactions is done before the transaction execution. Priority and time stamp rules are used to coordinate to determine the order of transaction execution to resolve concurrency conflicts. Second, reliable update algorithm is used to ensure the transmission reliability of transactions. Handle the transaction failure and do fault recovery. Last, do research on data synchronization conflicts and present complete data conflict resolution mechanism.The model of peer data synchronization based on node coordination strategy is used in the Network security monitoring system, according to the practical system requirements. This application implements the peer data synchronization of the critical data on the system, and solves the conflicts caused by the operating of the same critical data by multi-nodes. This application maintains data consistency.