数据中心光网络资源分配方法与优化

数据中心采用光网络传输模式有利于提高系统的灵活性和拓展性,进而降低整体的能耗水平,但在实践中发现光网络资源存在分配不合理、应用效率较低等问题,不利于数据中心应对流量增加、宽带资源紧张等挑战。通过机器学习建立流量分类算法模型,并将其设计为流量分类器,使其融入光网络的流量传输环节,实现流量自动分类。借助遗传算法模型重构光网络的拓扑结构,从而优化传输路径。通过以上两项措施改进光网络的资源分配方式,并借助软件模拟检验算法模型的性能,结果显示经过以上算法处理的光网络缩短了传输路径的长度,提高了网络资源的利用率。

Data Center Optical Network Resource Allocation Method and Optimization

The use of optical network transmission mode in data centers is beneficial to improve the flexibility and scalability of the system, and then reduce the overall energy consumption level. However, in practice, it is found that there are problems such as unreasonable allocation of optical network resources and low application efficiency, which is not conducive to the data center to cope with the challenges of traffic increase and broadband resource shortage. A traffic classification algorithm model is established by machine learning and designed as a traffic classifier, which is integrated into the traffic transmission link of optical network to realize automatic traffic classification. The genetic algorithm model is used to reconstruct the topology of optical network and optimize the transmission path. Through the above two measures to improve the allocation of optical network resources, and with the help of software simulation to check the performance of the algorithm model, the results show that the optical network processed by the above algorithm does reduce the length of the transmission path, improve the utilization of network resources.