臺灣博碩士論文加值系統

隨著現今網際網路的發展與網路技術的廣泛應用，傳統網際網路已從最早的區域科研網路和軍事網路演進成需要覆蓋語音串流、視訊串流、大量數據演算等的複雜型網路模式。過去的網路發展著重於設備性能的提升以及擴充，但隨著軟體定義網路(Software-defined Networking, SDN)與網路功能虛擬化(Network Functions Virtualization, NFV)的發表，近年來許多雲端服務供應商、電信服務供應商、甚至企業和學術網路都致力於布建軟體定義網路架構，從應用程式到基礎架構皆朝向靈活化、程式化、彈性化發展趨勢。軟體定義網路實現了以最少的硬體成本達到網路資源調度、彈性擴充的效果。本論文提出利用了SDN中控制平面獨立的特性，以一個控制器主導多部交換機，進而達到流量配置與負載平衡的效果。在伺服器和網路頻寬相異配置下，我們的SDN控制機制會在Openflow交換機上配置不同流表來分配由終端主機到伺服器的流量，進而使系統負載量均衡分配以及設備資源的平衡。

With the development and the extensive application of network technology, the Internet has evolved from the earliest regional scientific research and military network to a complex network model that needs to deal with voice streaming, video streaming, data calculation. In the past, network applications have been focusing on the improvement and expansion of the device capability, but since the release of Software-defined Networking (SDN) and Network Function Virtualization (NFV) technology, in recent years many Cloud and telecommunications service providers and companies have dedicated themselves to building SDN architectures. SDN allows realization of efficient and flexible network resource management and expansion with the lowest cost. This paper proposes to make use of the independent characteristics of the control plane in SDN, maneuvering multiple switches with one controller to achieve good traffic configuration and load balancing. First of all, The configuration of the corresponding requirements can be achieved in different Entry cases set up through virtual switches between end hosts with different load capacity, The configuration can be provisioned according to the differences in the server side. In presence of heterogeneous configuration of server capacity and network bandwidth, our SDN control mechanism can configure flow tables of Openflow switches of a network accordingly to allocate flows from the end hosts to the servers in order to balance the load distribution within a system with respect to the utilization of its equipment resources.

摘要 iii
ABSTRACT iv
誌謝 v
目錄 vi
圖目錄 viii
表目錄 ix
符號說明 x
一、緒論 1
1.1 研究動機 1
1.2 研究貢獻與論文架構 2
二、相關研究 3
2.1 軟體定義網路 3
2.2 SDN控制器 5
2.3 OpenFlow Switch與網路功能虛擬化技術 7
2.4 網路模擬器 9
三、系統模型與模擬方法 10
3.1 系統模型 10
3.2 模擬方法 15
3.3 Docker容器虛實整合 19
四、實驗模擬與分析 22
4.1 撰寫程式分析統計資料 22
4.2 限制網路頻寬下流量配置模擬實驗 25
4.3 硬體效能差異下流量配置模擬實驗 29
4.4 假設議題：多台控制器系統配置 31
五、結論 32
六、參考文獻 33
自傳 36

[1].Gunjan P Tank, Anmol Dixit, et al., 2017, “Software Defined Networks: The New Norm for Networks”, 3rd National Conference on "Recent Innovations in Science and Engineering", pp.33-35, Bengaluru, May.
[2].S. R. Chowdhury, M. F. Bari, et al., 2014, “PayLess: A low cost Network Monitoring Framework for Software Defined Networks”, IEEE/IFIP Network Operations and Management Symposium (NOMS), pp.1-9, Poland, May.
[3].Sushant Jain, Alok Kumar, et al., 2013, “B4: Experience with a Globally-Deployed Software Defined WAN” ACM SIGCOMM Computer Communication Review, Vol. 43, No. 4, pp.3-14, October.
[4].M. Kobayashi, S. Seetharaman, et al., 2014, “Maturing of OpenFlow and Software-defined Networking through deployments”, Elsevier B.V, vol. 61, pp.151-175, March.
[5].Bikash Koley, 2014, “Software Defined Networking at Scale”, on behalf of Google Technical Infrastructure.
[6].NTT DATA TAIWAN 官方網站，(2021)，https://www.nttdata-taiwan.com/ .
[7].Mahmoud Mohamed Abouzeid Mansour, Bao-Shuh Paul, 2016, Dynamic and Adaptive algorithm for load-balancing in SDN based Networks, National Chiao Tung University, master’s thesis.
[8].W. Braun, M. Menth, 2014, “Software-defined networking using OpenFlow: Protocols, applications and architectural design choices” Future Internet, vol. 6, pp.302-336, MDPI, May.
[9].Jen-Ting Chen, Jiun-Jian Liaw, 2020, Development of Forest Illegal Logging Monitoring System Based on SDN Wireless Mesh Network, Chaoyang University of Technology, master’s thesis.
[10].N. McKeown, T. Anderson, et al., 2008, “OpenFlow: enabling innovation in campus networks”, ACM SIGCOMN Computer Communication Review, Vol. 38, No. 2, pp. 69-74, April.
[11].Q. Mao, W. Shen, 2015, “A Load Balancing Method Based on SDN”, Seventh International Conference on Measuring Technology and Mechatronics Automation, Nanchang, China, June.
[12].Ping-Chi Chen, Wen-Ping Lai, 2016, Studies on Software Defined Video-Network Load Balancing Using Combined Multipath Resources, Yuan Ze University, master’s thesis.
[13].久保 類, 藤田 智成, 阿川 雄資, 2014，RYU-SDN framework，台灣資訊工業策進SDN團隊譯，台北。
[14].B. Pfaff, B. Davie, 2013, “The Open vSwitch Database Management Protocol”, RFC Editor, No. 7047, pp. 1-5, Internet Society (ISOC), December.
[15].Ta-Wei Yang, Discussion and Implementation for the High Availability of Firewall Virtualized Network Function on Openstack”, National Chiao Tung University, master’s thesis.
[16].IREAD01官方網站，(2019)，“Linux-虛擬網路裝置-Open vSwitch”, https://www.itread01.com/content/1548034955.html, January.
[17].EstiNet - Simulator | EstiNet11 官方網站，www.estinet.com .
[18].S. Wang, C. Chou, et al., 2013, “EstiNet OpenFlow Network Simulator and Emulator”, IEEE communication Magazine, Vol. 51, No. 9, pp.110-117, September.
[19].Docker/Docker Hub官方網站，(2021)，www.docker.com .
[20].軟體定義網路-EstiNet11實驗與應用，2020，思銳科技，台南。
[21].Ying-Dar Lin, Chin-Chiang Wang, et al., 2017, “Two-tier dynamic load balancing in SDN-enabled Wi-Fi networks”, springer Science+Business Media, New York, April.
[22].L. Deri, M. Martinelli, et al., 2014, “nDPI: Open-source high-speed deep packet inspection”, International Wireless Communications and Mobile Computing Conference(IWCMC), Nicosia, Cyprus, August.