簡易檢索 / 詳目顯示

研究生: 徐偉涵
Hsu, Wei-Hen
論文名稱: 在GENI上,實做與評估在軟體定義網路環境中的路由演算法
Implementation and Evaluation of Routing Algorithms in Software Defined Network on GENI
指導教授: 蔡明哲
Tsai, Ming-Jer
口試委員: 郭桐惟
張仕穎
學位類別: 碩士
Master
系所名稱:
論文出版年: 2018
畢業學年度: 106
語文別: 英文
論文頁數: 37
中文關鍵詞: 軟體定義網路路由演算法實驗床
外文關鍵詞: Software-Defined-Network, Routing-Algorithm, GENI
相關次數: 點閱:3下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 軟體定義網路越來越受到重視, 而許多的路由演算法相繼被提出, 評測演算法
    的表現能力成了重要的議題. 以實體的交換機搭建出的軟體定義網路環境雖然真實,
    但所需的成本極大. 實驗床是另一種搭建軟體定義網路環境的方法, 實驗床大部分
    是由某些組織或者企業所搭建出來的環境, 它提供給研究者搭建環境時所需要的計
    算及網路資源, 並且提供界面讓研究者可以在上面操作設定其所需的實驗環境, 因
    為所使用的資源是跟許多人共享使用的, 所以成本相較於實體交換機少很多, 而所
    量測出的實驗數據又比依靠軟體模擬可信許多.
    因此本論文中, 探討如何在一個GENI實驗床中建立出一個軟體定義網路, 並且
    我們將路由演算法區分為線下與線上的演算法, 針對這兩種演算法, 我們逐一探討
    如何將其放入所搭建好的軟體定義網路環境中並且進行測試與收集研究者所需要的
    測試結果.


    Software-Defined networking has received more and more attentation recently,
    lots of routing algorithms have been proposed. For evaluating the performace of a
    routing algorithm becomes a important issue. Although Using real OpenFlow switch
    for constructing a software-defined networking is quite real, the cost for building is
    vrey high. The network testbed is another way for build software-defined networking.
    The network testbed is usually created by some organizations or enterprises. It offers
    the compute and network resources for the researchers. Also, it offers a platform
    for the researchers to operate and build the experimental environment they wnat.
    Because the compute and network resources are shared by other researchers, the cost
    is lower compared to building by real machines. Also, the result of the experiment is
    more persuasive than testing by some simulators.

    中 文 摘 要 I Abstract II Contents III List of Figures VI 1 Introduction 1 2 General Background Information 3 2.1 Software-defined networking . . . . . . . . . . . . . . . 3 2.2 OpenFlow . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2.1 OpenFlow switch components . . . . . . . . . . 4 2.3 GENI . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.4 Open vSwitch . . . . . . . . . . . . . . . . . . . . . . . 6 2.5 Ryu controller . . . . . . . . . . . . . . . . . . . . . . . 6 3 System model 9 3.1 Workflow of GENI . . . . . . . . . . . . . . . . . . . . 9 3.1.1 Create a slice . . . . . . . . . . . . . . . . . . . 9 3.1.2 Create a topology . . . . . . . . . . . . . . . . . 9 3.1.3 Configure switches and controller . . . . . . . . 10 3.1.4 Run the experiment and collect the data . . . . 10 3.2 Network in GENI . . . . . . . . . . . . . . . . . . . . . 10 3.3 Ryu controller . . . . . . . . . . . . . . . . . . . . . . . 13 4 Implementation of Routing Algorithm in GENI 4.1 4.2 The reservation of compute resources for customized topol- ogy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Offline algorithm . . . . . . . . . . . . . . . . . . . . . 19 4.2.1 4.3 create the network graph for the input of routing algorithm . . . . . . . . . . . . . . . . . . . . . 19 4.2.2 setting up the demands . . . . . . . . . . . . . . 20 4.2.3 setting up the controller . . . . . . . . . . . . . 20 Online algorithm . . . . . . . . . . . . . . . . . . . . . 20 4.3.1 20 ryu controller . . . . . . . . . . . . . . . . . . . 5 Case study 5.1 5.2 16 22 Maximum Concurrent Flow Problem in MPLS-based Soft- ware Defined Networks . . . . . . . . . . . . . . . . . . 22 5.1.1 Problem definition . . . . . . . . . . . . . . . . 22 Algorithms of BFRMCF . . . . . . . . . . . . . . . . . 25 5.3 Implement method . . . . . . . . . . . . . . . . . . . . 25 5.3.1 The ryu controller . . . . . . . . . . . . . . . . 25 5.3.2 OpenFlw version and flow table size . . . . . . 25 5.3.3 iPerf . . . . . . . . . . . . . . . . . . . . . . . . 25 6 Experiment results 28 7 Conclusion 35 Bibliography 36

    [1] Open-Network-Foundation, https://www.opennetworking.org/.
    [2] GENI, http://www.geni.net/.
    [3] Open vSwitch, https://www.openvswitch.org/.
    [4] iPerf, https://iperf.fr/.
    [5] Tzu-Wen Chang, Yao-Jen Tang, Ming-Jer Tsai, Efficient Algorithm
    for Maximum Concurrent Flow Problem in MPLS-based Software
    Defined Networks.
    [6] N. Garg and J. Könemann. “Faster and simpler algorithms for
    multicom-modity flow and other fractional packing problems,” SIAM
    Journal on Computing, 2007.
    [7] A. Gushchin, A. Walid, and A. Tang, “Enabling service func-
    tion chaining through routing optimization in software defined net-
    works,” in IEEE Allerton Conference on Communication, Control,
    and Computing, 2015.
    36[8] S. Orlowski, R. Wessäly, M. Pióro, and A. Tomaszewski, “Sndlib
    1.0- survivable network design library,” Networks, 2010. [Online].
    Available: http://sndlib.zib.de

    QR CODE