Combined Degree-Based with Independent Dominating Set Approach for Controller Placement Problem in Software Defined Networks

2019 22nd Conference on Innovation in Clouds, Internet and Networks and Workshops (ICIN)(2019)

引用 7|浏览1
暂无评分
摘要
In Software Defined Networks (SDN), controllers should ideally be optimally placed to preserve the minimal response time between the controllers and the forwarding nodes in the data plane. It is well understood that the placement of controllers can greatly impact network performance in terms of controller responsiveness. An Independent Dominating Set (IDS) offers an optimized way of achieving the goal of minimal response time. However, finding an independent dominating set of minimal size is NP-hard, which means that there likely no polynomial-time algorithm can guarantee an optimal solution. In this paper, we propose a new node degree-based algorithm named High Degree with Independent Dominating Set (HDIDS) for the controller placement problem for the SDN networks. Our algorithm is composed of two phases to deal with controller placement: (1) determining candidate controller instances by selecting those nodes with highest node degree; and (2) partitioning the network into multiple domains, one controller per domain, exploiting the independent dominating set approach to ensure a distribution of controllers with lowest response times. Experiments results show that our algorithm, for a fixed number of selected controllers, has better performance in terms of minimizing the average response time between each controller and its forwarding nodes, and the response time of node with largest delay from the controller, as compared to a previously published algorithm based on optimized K-means. Further, we show that for an upper limit threshold for the required response time for an SDN, our algorithm nearly always requires fewer controllers to satisfy that threshold.
更多
查看译文
关键词
Node Degree,Independent Dominating Set,Controller placement,Response time,Software Defined Networks
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要