Protection And Recovery Method Based On Data Center Optical Network

With the gradual improvement of network technology, as well as the development of large data, the use of traditional single server cannot meet the existing data processing. In this case, more and more data centers are being put into use, the data center network traffic is also growing rapidly.In order to deal with a large number of data transmission, optical network with low latency, high bandwidth and other advantages, is widely Lused in data center networks. With the complication of data center optical inetwork,protection and recovery technology is paid more and more attention. It is a supporting teclhlology to guarantee network performance. In the case of network failure, protection and recover technology is used to realize the normal operation of the damaged service. Moreover, there are a lot of data in the optical network of the datacenter, in the event of failure, it will cause very serious losses, so the study of data center network protection and recovery technology is of great significance.The main work of this paper is around the data center optical network protection and recovery technology. After a comprehensive and systematic analysis of protection and restoration techniques for optical networks, the single node failure protection algorithm and the multi-node fault recovery algorithm are further studied. The main results are as follows:(1) Data center classification, planning, and classification were investigated and the optical network in the data center was analyzed in detail. Secondly, it summarizes and reviews the recovery and protection of optical network from the aspects of optical network protection and recovery strategy, network protection and recovery technology of each layer.(2) Through detailed analysis of optical network link fault protection technology, introducing its characteristics into node fault protection this paper proposes a data center optical network multicast protection algorithm based on single node failure, and through computer simulation, the results are analyzed in detail. The algorithm considers the shared path protection and link resource problem, and establishes the protection route for the destination nodes which are easily affected in the multicast transmission,so as to ensure that they can receive the data information transmission normally.(3) In the face of multi-data center node failure, a large-scale disaster recovery data center optical network algorithm to ensure that the recovery process to maximize the utilization of network data resources. The algorithm takes into account the acyclic hops of the network topology and the capacity sharing and backup of the data center nodes, and finds the optimal recovery order from the data center nodes of the fault. Through detailed simulation analysis, it is found that there is a good effect in the network connectivity, the mobility of data, the data rate of network fault and the probability of isolated.