海量数据环境下大型ISP网络流量卸载方法研究

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。

Huge amounts of data environment large ISP network traffic unloading method study

Huge amounts of data under the environment of large ISP network traffic explosive growth due to network congestion, the current network traffic unloading method through forecasting to uninstall the network traffic, uninstall the accuracy is low, poor quality of service.For this, put forward a new kind of huge amounts of data under the environment of large ISP network flow discharge method, through the maximum entropy method, based on the semantic information in the collection and traffic flow type for identification.Based on mass data environment of large ISP network traffic source node position relations, through the graph theory method to determine the best route of transmission, realize to unload large ISP network traffic.Based on the principles of relay node number and total discharge time at least, by Dijkstra method of mass data environment best uninstall route for solving large ISP network traffic.Experimental results show that the proposed method to unload large ISP network traffic, not only the flow type, high accuracy and high discharge rate, the high quality of service.