基於OpenFlow的網路負載平衡及高可用性系統

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：27

、訪客IP：3.129.195.254

姓名

洪源懋(Yuan-Mao Hong) 查詢紙本館藏

畢業系所

資訊工程學系

論文名稱

基於OpenFlow的網路負載平衡及高可用性系統
(OpenFlow-based Network Load Balancing and High Availability System)

相關論文

★ 無線行動隨意網路上穩定品質服務路由機制之研究	★ 應用多重移動式代理人之網路管理系統
★ 應用移動式代理人之網路協同防衛系統	★ 鏈路狀態資訊不確定下QoS路由之研究
★ 以訊務觀察法改善光突發交換技術之路徑建立效能	★ 感測網路與競局理論應用於舒適性空調之研究
★ 以搜尋樹為基礎之無線感測網路繞徑演算法	★ 基於無線感測網路之行動裝置輕型定位系統
★ 多媒體導覽玩具車	★ 以Smart Floor為基礎之導覽玩具車
★ 行動社群網路服務管理系統－應用於發展遲緩兒家庭	★ 具位置感知之穿戴式行動廣告系統
★ 調適性車載廣播	★ 車載網路上具預警能力之車輛碰撞避免機制
★ 應用於無線車載網路上之合作式交通資訊傳播機制以改善車輛擁塞	★ 智慧都市中應用車載網路以改善壅塞之調適性虛擬交通號誌

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 ( 永不開放)

摘要(中)

近幾年雲端運算（Cloud Computing）的興起，對ISP業者而言，來自使用者的流量與虛擬運算節點擴增的需求都大量的提升。但在現今的網路架構中，各設備皆獨具一格，其上有各自的作業系統與依附的應用程式，故難以因應現有網路架構下，在其上開發符合管理者需求的雲端應用程式。OpenFlow是作為實現具體實踐軟體定義網路（Software Defined Networking；SDN）願景的一開放式網路架構，目的是為了解決傳統交換器與路由器等網路設備因其作業系統與應用程式各自獨立，網路管理者無法因應自身需求來調整、管理網路資料的傳遞路徑等問題。故希望透過OpenFlow可集中式網管、快速佈署符合本身需求的應用程式等能力來增加網路封包路徑的彈性且可進一步將網路管理化繁為簡，提升網路的控制能力、降低網路的複雜度，並自行定義所需的網路。本論文提出智慧型的負載平衡演算法來克服不同flow entry其上負載不平衡的狀況來達到網路的負載平衡（Load balancing），並提供OpenFlow switch與伺服器的備援機制，透過OpenFlow來增加網路資料傳輸的彈性並達到高可用性的系統。
本論文中透過pre-configured flow與dynamic flow來使封包重導至指定的load balancing port上，並透過設置optimal netmask來降低流量項（Flow entry）的數量。另外，在OpenFlow controller上開發負載平衡演算法來決策分流的導向與重導資訊以達負載平衡效果，並利用health check機制來達到高可用性系統。最後，本論文在實驗結果中顯示所提之Genetic-based負載平衡演算法在一般情境會比Load-based及Hash-based負載平衡演算法針對系統負載程度平均後好上9.01%，更比Round-robin負載平衡演算法增進28.25%，顯示本機制更能夠在flow負載不均衡的情形下透過機制計算後進行重導，增進系統的負載平衡程度。

摘要(英)

As growth of the cloud computing, the traffic flow from clients and the demand for extending virtual nodes are also increasing. But it’s difficult to develop the network applications which match the requirement for network developer because the the applications and operating system for network devices are independent. OpenFlow is a solution for implementing Software Defined Networking(SDN) with the objective to improve the resilience of data transmission. Therefore, we use OpenFlow to achieve network control centrally, deploy applications fast and decrease the complexity for network topology. For these reasons, this thesis used OpenFlow to propose intelligent load balancing algorithms based on getetic algorithm to solve the situation for unbalanced load for each flow entries to achieve traffic load balancing and provide redundant mechanisms to build high availability system.
This thesis proposed pre-configured flow and dynamic flow load balancing mechanisms to redirect packet into specified load balancing port to achieve load balancing and set optional netmask to each pre-configured flows in order to reduce the flow rules. Furthermore, this thesis also developed four load balancing algorithms and two health check mechanisms on the OpenFlow controller. At last, the experiments show that genetic-based load balancing algorithm performs 9.01% degree of load balancing better than load-based and hash-based load balancing algorithm, and 28.25% better than round-robin load balancing algorithm. It shows that mechanisms proposed by this thesis can improve the degree of load balancing even the flow entries are unbalanced.

關鍵字(中)

★ 雲端運算
★ 軟體定義網路
★ OpenFlow
★ 負載平衡
★ 健康狀態檢測

關鍵字(英)

★ Cloud computing
★ Software defined networking
★ OpenFlow
★ Load balancing
★ Health check

論文目次

第1章緒論 1
1.1 概要 1
1.2 研究動機 3
1.3 研究目的 3
1.4 章節架構 4
第2章背景知識與相關研究 5
2.1 OpenFlow & OpenFlow vSwitch 5
2.2 OpenFlow controller 9
2.3 基因演算法 11
2.4 負載平衡 14
2.5 相關研究之比較 16
第3章研究方法 19
3.1 系統架構與設計 19
3.1.1. OpeFlow Handler模組 22
3.1.2. Load Balancing模組 23
3.1.3. Monitor模組 24
3.1.4. Decision模組 25
3.1.5. Policy Database模組 26
3.1.6. Flow control模組 28
3.1.7. Pre-configured flow模組 29
3.1.8. Dynamic flow模組 29
3.1.9. Session Maintenance模組 30
3.1.10. Health Check模組 31
3.1.11. Flap Dampening模組 32
3.2 系統運作流程與機制設計 33
3.2.1. 系統定義與假設 33
3.2.2. 資料符號表 34
3.2.3. 系統功能模組運作流程 37
3.2.4. Genetic-based負載平衡演算法運作流程 42
3.2.5. Pre-configured flow模組運作流程 47
3.2.6. Dynamic flow模組運作流程 48
3.2.7. Session Maintenance模組運作流程 51
3.2.8. Server Health Check模組運作流程 53
3.2.9. Flap Dampening模組運作流程 55
3.3 系統實作 58
第4章實驗與討論 64
4.1 情境一: Traffic information maintenance from controller 64
4.1.1. 實驗一：透過SNMP來擷取系統流量資訊 64
4.1.2. 實驗二：由OpenFlow的埠口統計資訊擷取系統流量資訊 66
4.2 情境二: OpenFlow switch health check 67
4.2.1. 實驗三：OpenFlow交換機的故障偵測與備援機制 67
4.3 情境三: Servers health check 71
4.3.1. 實驗四：Servers的故障偵測以及flap dampening機制運作 71
4.4 情境四: OpenFlow TCP load balancing 73
4.4.1. 實驗五：使用動態flows負載平衡並保持TCP連線不中斷 74
4.4.2. 實驗六：Dynamic flow with general difference value 77
4.4.3. 實驗七：Dynamic flow with significant difference value 79
4.5 情境五: Genetic-based負載平衡演算法 82
4.5.1. 實驗八： Fitness scaling in genetic algorithm 82
4.5.2. 實驗九：Fitness evaluation in genetic algorithm 83
4.6 情境六: OpenFlow UDP load balancing 83
4.6.1. 實驗十：Pre-configured flow with general difference value 84
4.6.2. 實驗十一：Pre-configured flow with significant difference value 97
4.6.3. 實驗十二：Flow entry的數量對負載平衡的影響 104
4.6.4. 實驗十三：演算法新增pre-configured flows的時間 105
4.6.5. 實驗十四：Short-term結合long-term負載平衡演算法 107
4.6.6. 實驗十五：使用多台虛擬機進行UDP負載平衡 110
第5章結論與未來研究方向 112
5.1 結論 112
5.2 研究限制 113
5.3 未來研究 113
參考文獻 115

參考文獻

[1] OpenFlow, http://www.OpenFlow.org/wp/learnmore/
[2] POX, www.noxrepo.org/pox/about-pox/
[3] Open vSwitch, http://openvswitch.org/
[4] Flow Visor, https://openflow.stanford.edu/display/DOCS/Flowvisor
[5] B. Sonkoly, A. Gulyas, F. Nemeth, J. Czentye, K. Kurucz, B. Novak and G. Vaszkun, ”OpenFlow Virtualization Framework with Advanced Capabilities,” Proceedings of the European Workshop on Software Defined Networking, pp.18-23, Oct. 2012.
[6] J. Matias, B. Tornero, A. Mendiola, E. Jacob, N. Toledo, ”Implementing Layer 2 Network Virtualization Using OpenFlow: Challenges and Solutions,” Proceedings of the European Workshop on Software Defined Networking, pp.30-35, Oct. 2012.
[7] E. Haleplidis, S. Denazis, O. Koufopavlou, J. Halpern, J.-H. Salim, ”Software-Defined Networking: Experimenting with the Control to Forwarding Plane Interface,” Proceedings of the European Workshop Software Defined Networking, pp. 91-96, Oct. 2012.
[8] P. Justin, G. Jesse, P. Ben, C. Martin, ”Virtual Switching in an Era of Advanced Edges,” Proceedings of the International Teletraffic Congress, Sep. 2011.
[9] Hui-Min Tseng, Hui-Lan Lee, Jen-Wei Hu, Te-Lung Liu, Jee-Gong Chang, Wei-Cheng Huang, ”Network Virtualization with Cloud Virtual Switch,” Proceedings of the IEEE 17th International Conference Parallel and Distributed Systems, pp.998-1003, Dec. 2011.
[10] Z. He, G. Liang, ”Research and Evaluation of Network Virtualization in Cloud Computing Environment,” Proceedings of the International Conference on Networking and Distributed Computing, pp.40-44, Oct. 2012.
[11] P. Yan, D. Yilong, A. Nakao, ”Cloud Rack: Enhanced virtual topology migration approach with Open vSwitch,” Proceedings of the International Conference on Information Networking, pp.160-164, Jan. 2011.
[12] Vu Tran Hoang, Pham Ngoc Nam; Tran Thanh, Le Thai Hung, Le Anh Van, Nguyen Duy Linh, To Duc Thien, Nguyen Huu Thanh, ”Power aware OpenFlow switch extension for energy saving in data centers,” Proceedings of the International Conference on Advanced Technologies for Communications, pp.309-313, Oct. 2012.
[13] O.E.-E. Ferkouss, R.-B. Ali, Y. Lemieux, C. Omar, ”Performance model for mapping processing tasks to OpenFlow switch resources,” Proceedings of the IEEE International Conference on Communications, pp.1476-1481, Jun. 2012.
[14] Beacon: A java-based openﬂow control platform. October 2011, http://www.beaconcontroller.net/
[15] N. Gude, T. Koponen, J. Pettit, B. Pfaff, M. Casado, N. McKeown, and S. Shenker, “Nox: Towards an operating system for networks,” ACM SIGCOMM Computer Communication Review, Jul. 2008.
[16] POX, www.noxrepo.org/pox/about-pox/
[17] Floodlight, http://www.projectfloodlight.org/floodlight/
[18] M. Jarschel, F. Lehrieder, Z. Magyari, R. Pries, “A Flexible OpenFlow-Controller Benchmark,” Proceedings of the European Workshop on Software Defined Networking, pp.48-53, Oct. 2012.
[19] H.-E. Egilmez, S.-T. Dane, K.-T. Bagci, A.-M. Tekalp, “OpenQoS: An OpenFlow controller design for multimedia delivery with end-to-end Quality of Service over Software-Defined Networks,” Proceedings of the Signal & Information Processing Association, pp.1-8, Dec. 2012.
[20] M. Seokhong, K. Seungju, L. Jaeyong, K. Byungchul, H. Wontaek, K. Jonguk, “Implementation of an OpenFlow network virtualization for multi-controller environment,” Proceedings of the International Conference on Advanced Communication Technology, pp.589-592, Feb. 2012.
[21] R. Bifulco, R. Canonico, M. Brunner, P. Hasselmeyer, F. Mir, “A Practical Experience in Designing an OpenFlow Controller,” Proceedings of the European Workshop on Software Defined Networking, pp.61-66, Oct. 2012.
[22] X. Zongyan, L. Haihua, G. Ye, “A Study on the Shortest Path Problem Based on Improved Genetic Algorithm,” Proceedings of the International Conference on Computational and Information Sciences, pp.325-328, Aug. 2012.
[23] Y. Minjie, “The Research of Path-Oriented Test Data Generation Based on a Mixed Ant Colony System Algorithm and Genetic Algorithm,” Proceedings of the International Conference on Wireless Communications, Networking and Mobile Computing, pp.1-4, Sep. 2012.
[24] k-means clustering, http://en.wikipedia.org/wiki/K-means_clustering
[25] L. Bin, J. Fangyuan, “An optimized genetic K-means clustering algorithm,” Proceedings of the International Conference on Computer Science and Information, pp.1296-1299, Aug. 2012.
[26] Back propagation neural network, http://en.wikipedia.org/wiki/Backpropagation
[27] C. Zhu, R. Rao, “The Improved BP Algorithm Based on MapReduce and Genetic Algorithm,” Proceedings of the International Conference on Computer Science and Service System, pp.1567-1570, Aug. 2012.
[28] L. Wentao, “SPGAO: A simulation platform for genetic algorithm based on OOP,” Proceedings of the International Conference on Automatic Control and Artificial Intelligence, pp.1910-1913, Mar. 2012.
[29] High Available, http://en.wikipedia.org/wiki/High_availability
[30] Service-Level Agreement , http://en.wikipedia.org/wiki/Service-level_agreement
[31] N. Handigol, S. Seetharaman, M. Flajslik, N. McKeown and R. Johari, “Plug-n-Serve: Load-balancing web trafﬁc using OpenFlow,” Demo at ACM SIGCOMM, Aug. 2009.
[32] R. Wang, D. Butnariu and J. Rexford, “OpenFlow-Based Server Load Balancing Gone Wild,” Proceedings of the 11th USENIX conference on Hot topics in management of internet, cloud, and enterprise networks and services, pp. 12-12, Mar. 2011.
[33] B. Radojevic and M. Zagar, “Analysis of issues with load balancing algorithms in hosted (cloud) environments,” Proceedings of the 34th International Convention MIPRO, pp.416-420, 23-27 May 2011.
[34] Daniel Stefani Marcon, Leonardo Richter Bays, “Flow Based Load Balancing: Optimizing Web Servers Resource Utilization,” Journals of Applied Computing Research, Dec. 2011.
[35] H. Uppal, D. Brandon, “OpenFlow Based Load Balancing,” University of Washington. CSE561: Networking. Project Report, Spring 2010.
[36] M. Koerner, O. Kao, “Multiple service load-balancing with OpenFlow,” Proceedings of the 13th International Conference on High Performance Switching and Routing, pp.210-214, 24-27 Jun. 2012.
[37] Coefficient of variation, http://en.wikipedia.org/wiki/Coefficient_of_variation
[38] V. Ramasubramanian, Emin G¨un Sirer, “The Design and Implementation of a Next Generation Name Service for the Internet,” Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communications, pp.331-342, Oct. 2004.
[39] Pica8 Open Switch, http://www.pica8.com/open-switching/open-switching-overview.php
[40] Consistent hash, http://en.wikipedia.org/wiki/Consistent_hashing
[41] MD5 hash, http://en.wikipedia.org/wiki/MD5
[42] Intrusion detection system, http://en.wikipedia.org/wiki/Intrusion_detection_system
[43] Distributed denial-of-service attack, https://en.wikipedia.org/wiki/Denial-of-service_attack

指導教授

周立德(Li-Der Chou)

審核日期

2013-8-28

推文