面向一类新型数据中心网络的容错与安全性能研究

随着5G通信与云计算的快速发展，数据中心网络直接影响到5G与云计算所提供的服务质量与安全问题。一类以服务器为中心的新型数据中心网络被提出以适应新型应用的需求。随着网络规模扩大，网络容错性与安全性变得愈发重要。现有的评估网络容错性方法大都是基于传统条件，容错性指标抗攻击能力较差，存在安全性和实时性较低的问题。针对上述问题，本项目拟通过总结这类网络的共同性质，使用网络结点隐私保护，网络连通性与容错路由来提高数据中心网络的容错与安全性能。项目具体内容包括：1）设计性能优越的网络结点隐私保护算法，以提高网络的容错性；2）构建高效且复杂度较低的拓扑重构算法，降低网络重构的资源开销；3）构造长度接近于直径的最短路径容错路由算法，以保证通信在最大拥塞度方面具有高效的性能；4）分析算法的时间复杂度并进行实验验证与优化。以上研究对于这类网络的实际应用与推广具有重要的理论意义和实际价值。

With the rapid development of 5G communication and cloud computing, data center networks directly affect the quality of services and security. A novel class of server-centric data center network is proposed to meet the requirements of new applications. With the expansion of network scale, network fault tolerance and security become more and more significant. The existing methods for evaluating network fault tolerance are mostly based on traditional conditions, and fault tolerance indicators have poor resistance to attacks. There are problems of low security and real-time performance. To address the above-mentioned issues, this project intends to improve the fault tolerance and security performance of data center networks by summarizing the common nature of such networks, using network node privacy protection, network connectivity and fault-tolerant routing. The specific content of the project includes: 1) design a privacy protection algorithm for network nodes with superior performance to improve the fault tolerance of the network; 2) build a topology reconstruction algorithm with superior performance to reduce the resource overhead of network reconstruction; 3) construct the shortest path fault-tolerant routing algorithm whose length is close to the diameter to ensure that the communication has efficient performance in terms of maximum congestion; 4) analyze the time complexity of the algorithm and conduct experimental verification and optimization. The above research has important theoretical significance and practical value for the practical application and promotion of such networks.