摘要
如何在规模庞大的数据中心网络内部进行负载均衡,实现网络资源的优化利用,是个亟待解决的问题.OpenFlow技术提供了一种可编程操作的流量控制方法,已成为近期解决负载均衡问题的热门关键技术.针对数据中心常见的Fat-tree网络拓扑,就数据流在OpenFlow网络中的路径设计问题,利用OpenFlow网络控制器集中控制的优势,在控制器上设计并实现了一种动态负载均衡路由策略.该策略首先选出源节点与目的节点间的前K条最短路径,然后根据获取的网络链路实时状态信息,对选出的前K条路径进行多指标综合评价,并从中选出数据流的最优转发路径.最后,采用Floodlight控制器和Mininet仿真平台,对提出的负载均衡路由策略进行可行性验证和性能验证.结果表明,该策略有效缓解了网络阻塞,提高了链路利用率.
The OpenFlow technology which provides a programmable flow control method has become the hot key technology to solve the load balancing problem in networks. In this thesis, a dynamic load balancing muting algorithm about choosing data transportation paths in Fat-tree network commonly used in DC （ data center ）, was proposed and implemented on the controller D utilizing the centralized controlling advantages of controllers in OpenFlow networks. The algorithm would choose the K shortest paths among all the paths between the source node and the destination node ｜ the final best forwarding path would be determined by the former K paths ＇ multi-index comprehensive evaluation values which were calculated based on the collected network status information in real time. Finally, the Floodlight controller and Mininet simulation was used in verifying the feasibility and performance of the load balancing routing algo- rithm proposed in this paper. The results show that the load balancing muting strategy can effectively alleviate the congestion of networks and improve the utilization of links.
出处
《小型微型计算机系统》
CSCD
北大核心
2017年第2期209-212,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61472192)资助
