隨著網路技術的快速發展，以分散式架構的傳統網路導致網路複雜度上升，同時也使維護的成本增加。為了提升網路應用服務品質，軟體定義網路(Software-Defined Networking, SDN)技術因應而生，以符合現今網路服務多樣化之需求。目前SDN已具許多成功應用及技術實例，其可程式化的特性，也使得網路虛擬化(Network Virtualization)的概念因應而生，使得網路資源得以透過虛擬化方式共享。因此，本研究聚焦於網路虛擬化之資源管理機制進行研究，透過軟體自訂義之方式實現網路流量分配機制，達到於網路切片(Slices)中皆能有效運用網路資源，並提高使用者最佳服務品質目標。

本研究致力於SDN網路虛擬化之網路資源分配機制研究，透過網路虛擬化Slicing技術，將實體網路切成多個虛擬網路，並於虛擬網路中，針對各種不同的網路切片進行效能分析，進而規劃出有效的網路資源分配方式，使得各網路切片可依據當前網路狀況進行網路資源調整，解決各網路切片因規劃不完善所造成的瓶頸，以達到最佳網路資源運用，同時提供使用者高品質服務。

本研究提出企業網路虛擬化平台與資源分配機制，並針對所提出的系統進行效能分析。首先，本機制將於各網路切片資源利用率進行監控與分析，並針對各網路切片之網路資源進行供給與需求評估。而後，本機制針對各網路切片之網路資源分配進行成本評估，以擬定最小成本之網路資源分配規劃。最後，依據網路資源分配規劃變更網路資源分配情形。根據研究結果顯示，本研究所提出的分配機制與未使用本機制之Slicing技術相互比較下，有效解決各網路切片之間的瓶頸，並提升網路資源利用率13.4%，以及相較於未使用任何Slicing技術，本研究機制將有效提升網路資源利用率25.7%。

Along with the rapid development of the Internet, the distributed architecture of legacy networks has increased network complexity and the cost of maintenance. Traditional network cannot meet the requirements of current users. To meet the requirements and improve the quality of various network application services, Software-Defined Networking (SDN) has been developed.

SDN technology has already been used successfully in many cases. In SDN architecture, network intelligence and state are logically centralized, and the underlying network infrastructure is abstracted from the applications. As a result, SDN provides the benefits of programmability, automation, and network control it. It also realizes the concept of network virtualization, through which network resources can be shared. Therefore, this work concerns the allocation of network resources on a network virtualization platform. This work presents a dynamic resource allocation mechanism in EnterpriseVisor based on the proposed network virtualization platform, to deploy the network resources based on an analysis of the resource utilization of each network slice.

This work analyzes the proposed dynamic resource allocation mechanism in EnterpriseVisor. First, the EnterpriseVisor monitors the resource utilization of each slice. Second, based on the results, the mechanism of EnterpriseVisor analyzes the utilization of each slice. Then, the slices will be classified as providers or requesters. Finally, the EnterpriseVisor will perform resource planning by minimizing cost and translating into resource allocation policies. Based on the results thus obtained, the proposed EnterpriseVisor overcomes the bottleneck of slicing technology and improves resource utilization by 13.4% and 25.7% over those achieved SDN using the slicing technology and SDN without slicing technology.

[1] S. Shah, J. Faiz, M. Farooq, A. Shafi and S. Mehdi, “An Architectural Evaluation of SDN Controllers,” Proceedings of the IEEE International Conference on Communications, pp. 3504-3508, 2013.
[2] S. Sezer, S. Scott-Hayward, P. Chouhan, B. Fraser, D. Lake, J. Finnegan, N. Viljoen, M. Miller and N. Rao, “Are We Ready for SDN? Implementation Challenges for Software-defined Networks, ” IEEE Communications Magazine, vol. 51, no. 7, pp. 36-43, 2013.
[3] A. Dixit, F. Hao, S. Mukherjee, T. Lakshman and R. Kompella, “Towards an Elastic Distributed SDN Controller,” Proceedings of the ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking, pp. 1-6, 2013.
[4] S. Ortiz, “Software-Defined Networking: On the Verge of a Breakthrough?, ” Computer, vol. 46, no. 7, pp. 10-12, 2013.
[5] M. Karl, J. Gruen and T. Herfet, “Multimedia Optimized Routing in OpenFlow Networks,” Proceedings of the IEEE International Conference on Networks, pp. 11-13, 2013.
[6] A. Tootoonchian and Y. Ganjali, “HyperFlow: A Distributed Control Plane for OpenFlow,” Proceedings of the Internet Network Management Conference on Research on Enterprise Networking, pp. 3-3, 2010.
[7] Y. Chen, X. Gong, W. Wang and X. Que, “VNMC for Network Virtualization in OpenFlow Network,” Proceedings of IEEE International Conference on Cloud Computing and Intelligent Systems, pp. 797-801, 2012.
[8] B. Sonkoly and A. Gulyas, “Integrated OpenFlow Virtualization Framework with Flexible Data, Control and Management Functions,” Proceedings of the IEEE INFOCOM, pp. 13-14, 2012.
[9] D. Turull, M. Hidell and P. Sjodin, “LibNetVirt: The Network Virtualization Library,” Proceedings of the IEEE International Conference on Communications, pp. 5543-5547, 2012.
[10] R. Sherwood, G. Gibb, K. Yap, G. Appenzeller, M. Casado, N. McKeown and G. Parulkar, “FlowVisor: A Network Virtualization Layer,” Technical Report Openflow, 2009.
[11] M. Kim, J. Lee, B. Kim and W. Hong, “Implementation of an OpenFlow Network Virtualization for Multi-Controller Environment,” Proceedings of the International Conference on Advanced Communication Technology, pp. 589-592, 2012.
[12] W. Wang , Y. Hu, X. Que and G. Xiangyang, “Autonomicity Design in OpenFlow Based Software Defined Networking,” Proceedings of the IEEE International Workshop on Management of Emerging Networks and Services, pp. 818-823, 2012.
[13] M. Fernandez, “Comparing OpenFlow Controller Paradigms Scalability: Reactive and Proactive,” Proceedings of the IEEE International Conference on Advanced Information Networking and Applications, pp. 1009-1016, 2013.
[14] M. Shirazipour, W. John, J. Kempf, H. Green and M. Tatipamula, “Realizing Packet-optical Integration with SDN and OpenFlow 1.1 Extensions,” Proceedings of the IEEE International Conference on Communications, pp. 10-15, 2011.
[15] A. Furculita, M. Ulinic, A. Rus and V. Dobrota, “Implementation Issues for Modified Dijkstra's and Floyd-Warshall Algorithms in OpenFlow,” Proceedings of the RoEduNet International Conference on Networking in Education and Research, pp. 26-28, 2013.
[16] A. Vishnoi, R. Poddar, V. Mann and S. Bhattacharya, “Effective Switch Memory Management in OpenFlow Networks,” Proceedings of the ACM International Conference on Distributed Event-Based Systems, pp. 177-188, 2014.
[17] T. Fratczak, M. Broadbent, P. Georgopoulos and N. Race, “HomeVisor: Adapting Home Network Environments,” Proceedings of the European Workshop on Software Defined Networks, pp. 32-37, 2013.
[18] M. Koerner and O. Kao, “Multiple Service Load-Balancing with OpenFlow,” Proceedings of the IEEE International Conference on High Performance Switching and Routing, pp. 210-214, 2012.
[19] E. Salvadori, R. Corin, A. Broglio and M. Gerola, “Generalizing Virtual Network Topologies in OpenFlow-based Networks,” Proceedings of the IEEE Global Telecommunications Conference, pp. 1-6, 2011.
[20] R. Corin, M. Gerola, R. Riggio, F. Pellegrini and E. Salvadori, “VeRTIGO: Network Virtualization and Beyond,” Proceedings of European Workshop on Software Defined Networking, pp. 24-29, 2012.
[21] K. Yap, M. Kobayashi, R. Sherwood, N. Handigol, T. Huang, M. Chan and N. McKeown, “OpenRoads: Empowering Research in Mobile Networks,” Proceedings of the ACM SIGCOMM, pp. 3-4, 2009.
[22] N. Handigol, S. Seetharaman, M. Flajslik, N. McKeown and R. Johari, “Plug-n-Serve: Load-Balancing Web Traffic Using OpenFlow,” Proceedings of the ACM SIGCOMM, pp. 23-24, 2009.
[23] G. Gibb, D. Underhill, A. Covington, T. Yabe and N. McKeown, “OpenPipes: Prototyping High-speed Networking Systems,” Proceedings of the ACM SIGCOMM Conference, pp. 19-20, 2009.