Adaptive Distributed Software Defined Networking

Adaptive Distributed Software Defined Networking
复制标题

自适应分布式软件定义网络

DOI:
10.1016/j.comcom.2016.11.009
复制
发表时间:
2017-04
期刊:
Elsevier Computer Communications
影响因子:
--
通讯作者:
Yong Jiang
Yong Jiang
中科院分区:
其他
文献类型:
--
作者:
Yanyu Chen;Yuan Yang;Qi Li;Yong Jiang

文献摘要

参考文献

相似文献

分布式软件定义网络(SDN)将网络中的多个控制器联合起来,以解决单控制器网络中的问题,例如,提高网络可靠性,减少控制器与交换机之间的延迟。然而,在目前的分布式SDN方案中,SDN交换机和控制器之间的映射是静态配置的,这可能会导致控制器之间的负载分配不均匀。这些方案不能完全受益于分布式SDN体系结构。针对这一问题,本文提出了一种自适应弹性分布式SDN体系结构ESDN。该体系结构动态地选择交换机连接到的最小数量的活动控制器,并根据网络负载改变交换机和控制器之间的映射。特别是,交换机可以从一个控制器域迁移到另一个控制器域,以便映射适应网络负载。我们将控制器选择问题形式化为一个优化问题,并证明了该问题是NP难的。我们分别使用离线算法和在线算法来解决这个问题。利用启发式算法,网络中的控制器相对于网络负载动态改变。离线算法对最优解的逼近比为2,在线算法可以在较短的时间内找到相似数目的主动控制器。通过仿真验证了算法的有效性,并对算法的性能进行了评估。特别是,当整个网络负载从65%的控制器容量下降到25%的控制器容量时,通过在线算法的收缩作用计算的非活动控制器数平均达到最优值的92%左右。
Distributed Software Defined Networking (SDN) federates multiple controllers in a network to solve the problems in single controller networks, e.g., to improve network reliability and reduce the delay between controllers and switches. However, in the current distributed SDN schemes, the mapping between SDN switches and controllers is statically configured, which may result in uneven load distribution among controllers. These schemes cannot fully benefit from the distributed SDN architecture. In order to address this issue, this paper proposes ESDN, anadaptive elastic distributed SDN architecture. The architecture dynamically selects a minimum number of active controllers that switches attached to, and changes the mapping between switches and controllers according to the network load. Specially, a switch can migrate from one controller domain to another so that the mapping is adaptive to the network load. We formalize the controller selection problem as an optimization problem, and prove that the problem is NP-Hard. We solve the problem by using offline and online algorithms, respectively. With the heuristics, controllers in a network are dynamically changed with respect to the network load. The offline algorithm has an approximation ratio of 2 related to the optimal result, and the online algorithms can find similar number of active controllers within a shorter time. We validate the algorithms and evaluate the performance by simulations. In particular, the number of inactive controllers computed by shrinking action of online algorithm averagely achieves around 92% of the optimal values when the whole network load decreases from 65% controller capacity to 25% controller capacity.
DOI: 10.1145/2535771.2535789
发表时间: 2013-11
期刊: Proceedings of the Twelfth ACM Workshop on Hot Topics in Networks
影响因子: --
作者:
Matthew Monaco;Oliver Michel;Eric Keller
通讯作者: Matthew Monaco;Oliver Michel;Eric Keller
DOI: 10.1145/2491185.2491198
发表时间: 2013-08
期刊: --
影响因子: --
作者:
S. Schmid;J. Suomela
通讯作者: S. Schmid;J. Suomela
DOI: --
发表时间: --
期刊: --
影响因子: --
作者:
通讯作者: --
DOI: 10.1145/2491185.2491189
发表时间: 2013-08
期刊: --
影响因子: --
作者:
David Erickson
通讯作者: David Erickson
DOI: 10.1145/2491185.2491194
发表时间: 2013-08
期刊: --
影响因子: --
作者:
Laurent Vanbever;Joshua Reich;Theophilus A. Benson;Nate Foster;J. Rexford
通讯作者: Laurent Vanbever;Joshua Reich;Theophilus A. Benson;Nate Foster;J. Rexford