臺灣博碩士論文加值系統

如今軟體定義網路越來越流行相關的研究也越來越多，因此它被應用於各種網路環境中。雖然許多論文研究已經闡述了軟體定義網路的幾個好處，但是軟體定義網路部署帶外模式的主要問題是控制器的成本昂貴，尤其在大型網路拓撲和多控制器的部署方式，用於軟體定義網路部署的帶內模式並未引起太多的關注。
在本文中，研究帶內模式軟體定義網路部署多個控制器，再利用基於圖形的k-Means Clustering演算法去找出多個控制器最佳的部署位置,目標以達到各個裝置到控制器的響應時間最短。
根據Topology Zoo和IMC提供的數據中心資料集，我們使用Mininet建構帶內模式軟體定義網路部署多個控制器。分析顯示除了解決成本問題外，正確的部署多個控制器還可以有效的降低裝置到控制器的響應時間。實驗結果還表明，根據拓撲的分布情況有時候會出現控制器飽和的現象,就算持續部署控制器也不會得到響應時間縮短。所以並不是部署越多的控制器就能得到一定比率的響應時間縮短，還有帶內模式下控制器位置是至關重要的，位置不正確也無法降低響應時間。

Nowadays Software-Defined Networks(SDN) paradigm becomes more popular in computer networking society, therefore it is applied in various networked environments. Although the literature has articulated several benefits of this approach in terms of improved flexibility and performance, the major problem of Out-of-Band mode for Software-Defined Networks (SDN) deployment is the expensive cost of the controller, especially in large network topology and multiple controller deployments. However, the multiple controllers of In-Band mode for Software-Defined Networks (SDN) deployments have not received much attention. In this thesis, we study In-Band mode for Software-Defined Networks (SDN) deployment of multiple controllers, using Graph-Based k-Means Clustering Algorithm to find out the best position for multiple controllers to achieve the shortest response time to each device.
With the data set from the IMC data center and Topology Zoo, the simulation of In-Band mode for SDN deployment multiple controllers is conducted in Mininet emulator. The analysis shows a significant reduction in not only cost but also effectively the response time of the device to the controller. However, the more controllers, response time won''t always be shortened. It’s based on topology distribution, the deployment sometimes will response time convergence phenomenon.

Contents
List of Figures ii
List of Tables iv
1 Introduction 1
2 Related Works 5
2.1 Out-of-Band versus In-Band .............................................................................. 5
2.1.1 Out-of-Band control plane ....................................................................... 6
2.1.2 In-Band control plane .............................................................................. 6
2.2 Overview of SDN with multiple controllers....................................................... 7
2.3 Controller placement in SDN of In-Band mode ............................................... 10
3 Controller Deployment Strategy 12
3.1 K-Means Clustering Algorithm ........................................................................ 12
3.2 Graph-Based k-Means Clustering Algorithm ................................................... 15
3.3 Deployment Strategy of Multiple SDN Controllers ......................................... 16
4 Simulation environment and Parameter setting 22
5 Simulation Results 31
6 Conclusions 39
Bibliography 42

[1] N. McKeown et al., "OpenFlow: enabling innovation in campus networks," ACM SIGCOMM Computer Communication Review, vol. 38, no. 2, pp. 69-74, 2008.
[2] T. Koponen et al., "Onix: A Distributed Control Platform for Large-scale Production Networks," in Operating Systems Design and Implementation, vol. 10, pp. 1-6, 2010.
[3] A. Tootoonchian and Y. Ganjali, "HyperFlow: A distributed control plane for OpenFlow," in Proceedings of the Internet Network Management Conference on Research on Enterprise Networking, pp. 3-3, 2010.
[4] S. Hassas Yeganeh and Y. Ganjali, "Kandoo: a framework for efficient and scalable offloading of control applications," in ACM Proceedings of the first workshop on Hot Topics in Software Defined Networks, pp. 19-24, 2012
[5] B. Heller, R. Sherwood, and N. McKeown, “The controller placement problem,” in Proceedings Hot Topics in Software Defined Networks, pp. 7–12, 2012.
[6] G. Yao, J. Bi, Y. Li, and L. Guo, "On the capacitated controller placement problem in software-defined networks,” IEEE, Communications Letters, vol. 18, no. 8, pp. 1339–1342, 2014.
[7] A. Sallahi and M. St-Hilaire, “Optimal Model for the Controller Placement Problem in Software Defined Networks,” IEEE, Communications Letters, vol. 19, no. 1, pp. 30–33, 2015.
[8] U. Krishnaswamy et al., “ONOS: An open source distributed SDN OS,” [Online]http://www.slideshare.net/ umeshkrishnaswamy/open-networkoperating-system, 2013.
[9] A. Dixit, F. Hao, S. Mukherjee, T. V. Lakshman, and R. Kompella, “Towards an elastic distributed SDN controller,” in Proceedings of the 2nd ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking (HotSDN ''13), pp. 7–12, Hong Kong, 2013.
[10] Y. Fu, J. Bi, K. Gao, Z. Chen, J. Wu, and B. Hao, “Orion: a hybrid hierarchical control plane of software-defined networking for large-scale networks,” in Proceedings of the 22nd IEEE International Conference on Network Protocols (ICNP ''14), pp. 569–576, Raleigh, NC, USA, 2014.
[11] S. Khuller and Y. J. Sussmann, “The capacitated k-center problem,” Society for Industrial and Applied Mathematics (SIAM) Journal on Discrete Mathematics, vol. 13, no. 3, pp. 403–418, 2000.
[12] Y. Zhang, N. Beheshti, and M. Tatipamula, “On resilience of split-architecture networks,” in IEEE Global Telecommunications Conference (GLOBECOM 2011), pp. 1–6, 2011.
[13] Y.-n. HU, W.-d. WANG, X.-y. GONG, X.-r. QUE, and S.-d. CHENG, “On the placement of controllers in software-defined networks,” The Journal of China Universities of Posts and Telecommunications, vol. 19, pp. 92–171, 2012.
[14] L. F. M¨uller, R. R. Oliveira, M. C. Luizelli, L. P. Gaspary, and M. P. Barcellos, “Survivor: an Enhanced Controller Placement Strategy for Improving SDN Survivability,” in IEEE Global Communications Conference (GLOBECOM), 2014.
[15] Y. A. Jim´enez Agudelo, C. Cervell´o Pastor, A. J. Garc´ıa., “Defining a network management architecture,” in Proc. 21st IEEE International Conference on Network Protocols (ICNP), 2013.
[16] Y. Jimenez, C. Cervello-Pastor, and A. J. Garcia, “On the controller placement for designing a distributed SDN control layer,” in IEEE Networking Conference, Information Security Conference and Privacy Conference (IFIP), pp. 1–9, 2014.
[17] A. Ksentini, M. Bagaa, T. Taleb, and I. Balasingham, “On using bargaining game for optimal placement of sdn controllers,” in Proc. IEEE International Conference on Communications, 2016.
[18] H. K. Al-Mohair, J. M. Saleh, and S. A. Suandi, “Hybrid Human Skin Detection Using Neural Network and K-Means Clustering Technique,” Applied Soft Computing, vol. 33, pp. 337–347, 2015.
[19] C. Jose, L. A. R. Calla et al., “Parallelization of the Algorithm K-means Applied in Image Segmentation,” International Journal of Computer Applications, vol. 88, no. 17, 2014.
[20] J. A. Hartigan and M. A. Wong ”A K-Means Clustering Algorithm” In Proceedings of Journal of the Royal Statistical Society. Series C (Applied Statistics), Vol. 28, No. 1, pp. 100-108, 1979.
[21] “K-means clustering”, https://en.wikipedia.org/wiki/K-means_clustering
[22] Theophilus Benson, Aditya Akella and David A. Maltz, “Network Traffic Characteristics of Data Centers in the Wild”, Proceedings of the 10th ACM SIGCOMM conference on Internet measurement Pages 267-280, 2010.
[23] S. Knight, H. Nguyen, N. Falkner, R. Bowden, and M. Roughan. “The internet topology zoo”, Proceedings of the IEEE Journal on Selected Areas in Communications, 2011
[24] M. Ferrer, E. Valveny, F. Serratosa, I. Bardají, and H. Bunke, “Graph-based k-means clustering: A comparison of the set median versus the generalized median graph,” in Proc. 13th Computer Analysis of Images and Patterns (CAIP), pp. 342–350, 2009.