海量数据环境下大型ISP网络流量卸载方法研究

发布时间：2018-02-06 07:47

  本文关键词： 海量数据环境 大型ISP网络 流量卸载　出处：《科学技术与工程》2017年13期 　论文类型：期刊论文

【摘要】：海量数据环境下大型ISP网络流量爆炸性增长造成网络阻塞。当前网络流量卸载方法通过预测确定待卸载网络流量,卸载准确率较低,服务质量差。为此,提出一种新的海量数据环境下大型ISP网络流量卸载方法,通过最大熵法,依据采集流量中的语义信息对流量类型进行识别。依据海量数据环境下大型ISP网络流量源节点的位置关系,通过图论法确定最佳传输路线,实现对大型ISP网络流量的卸载。依据中继节点数量与总卸载时间最少原则,通过Dijkstra方法对海量数据环境下大型ISP网络流量最佳卸载路线进行求解。实验结果表明,采用所提方法对大型ISP网络流量进行卸载,不仅流量类型识别精度高,而且卸载率高,服务质量高。
[Abstract]:The explosive growth of large ISP network traffic in massive data environment results in network congestion. Current network traffic unloading methods determine the network traffic to be unloaded by prediction. The unload accuracy is low and the quality of service is poor. In this paper, a new method of unloading large ISP network traffic under the environment of massive data is proposed, and the maximum entropy method is adopted. According to the semantic information of the collected traffic, the traffic type is identified. According to the location relationship of the traffic source node in the large-scale ISP network under the massive data environment, the best transmission route is determined by the graph theory method. Realize the unloading of large ISP network traffic, according to the number of relay nodes and the principle of minimum total unload time. The Dijkstra method is used to solve the optimal unloading route of large scale ISP network traffic under the environment of massive data. The experimental results show that the proposed method is used to unload the large ISP network traffic. Not only the accuracy of traffic type identification is high, but also the unloading rate is high, and the quality of service is high.
【作者单位】： 百色学院信息工程学院;
【基金】：2015年广西高校科学技术研究项目(KY2015ZD118)资助
【分类号】：TP393.06
，

本文编号：1493952