Study On Survivability Of Large Scale Faults In Optical Networks

With the continuous maturity of optical network technology and the expansion of its application scope,optical network is more and more widely used in our daily life.Optical network has the characteristics of large capacity,complex structure and decentralized geographical location,especially the widespread application in data center network.These characteristics make optical network itself face a series of challenges.For example,earthquakes,tornadoes,tsunamis and other natural disasters or human error may lead to regional large-scale failures of optical networks,which will have a great impact on data service transmission.Therefore,the survivability of optical networks becomes particularly important.In the large-scale network failure,theresearch of reasonable network recovery strategy has always been one of the key issues in the field of network communication.This paper focuses on the problem of large-scale fault survivability in optical networks,and focuses on the following work:(1)This paper analyzes and discusses the development status,survivability status,key technologies,fault switching process,protection and recovery principles and other basic theoretical knowledge of optical networks.(2)This paper introduces the existing node importance ranking algorithm,analyzes its principle and advantages and disadvantages.Because the traditional K-shell algorithm has defects in the node importance ranking problem,this paper introduces a variety of node importance metrics,and proposes a node importance ranking algorithm combining multiple index entropy weight and K-shell to solve the key node ranking problem.(3)In this paper,the proposed algorithm is used to solve the key node sorting problem,create a node candidate set according to the proposed algorithm strategy,and find its recovery order.By using the criteria of multiple indicators to evaluate the quality of network recovery,and comparing with other recovery algorithms through simulation analysis,the feasibility of this algorithm is obtained.(4)By combining the EWKB algorithm and the service routing strategy proposed in this chapter,we can plan the traffic transmission in the network and establish a large-scale failure scenario.By comparing with the Dijkstra algorithm,we can recover the nodes one by one in the routing planning.The advantages of this kind of service routing allocation in survivability Solve the paralysis problem of large-scale optical network as quickly as possible and minimize the loss caused by failure(5)This paper introduces the related technologies and functions of the multi-layer and multi domain optical network planning system software jointly developed by myself and my classmates in the laboratory.The survivability analysis module of the system is analyzed and simulated in detail.This software is an intelligent and professional communication system network simulation development and operation platform,which can provide strong support for optical network design and planning,and provide the function of invulnerability reconfiguration capability analysis in the scene of large-scale failures in optical networks.