軟體定義網路（Software Defined Networking, SDN）為新的網路架構，此架構提供中央控制的能力，讓管理員能快速且動態的進行資料流（Data Flow）的控制。OpenFlow為此架構下常用的協定，此協定可讓管理員加入多個比對欄位的流量條目（Flow Entry）至交換機的流量表（Flow Table）中來精準的控制資料流。為了要快速的比對這些比對欄位較多的流量條目，支援OpenFlow的交換機會採用三態內容定址式記憶體（Ternary Content Addressable Memory, TCAM）來儲存流量條目。然而，三態內容定址式記憶體受限於電路設計上的考量，使其容量不足以管理資料中心的資料流，而造成了網路效能的衰減。為此，本研究提出了流量條目代理機制（Flow Entry Agent），此代理機制會透過異質性流量表整合（Heterogeneous Tables Integration）、流量表分割與化簡（Table Division and Reduction）以及虛擬Packet In（Pseudo Packet In）三個機制來提升OpenFlow交換機中可儲存的流量條目數量。經由模擬顯示，我們的機制讓交換機內可儲存的流量條目數量成長為原先的6到18倍。

Software Defined Networking (SDN) is a new network architecture, which provides centralized controllability for network administrators to do fast and dynamic flow control. OpenFlow, the most commonly used protocol in SDN, an administrator can control data flows precisely by multiple matching fields of flow entries in flow tables. In order to perform high-speed flow matching process with these entries, commercial OpenFlow-enabled legacy switches implement flow tables with Ternary Content Addressable Memory (TCAM) to store flow entries. However, the size of TCAM-based flow table is restricted to few thousands of entries due to manufacturing cost and high power consumption. The limited size of TCAM-based table is not sufficient to manage the data flows in a data center network, which decreases the performance of OpenFlow network. Therefore, we propose Flow Entry Agent in this thesis. The Agent include three mechanisms, Heterogeneous Tables Integration, Table Division and Reduction and Pseudo Packet In. These mechanisms can increase the number of flow entries that can be stored in a switch. Our simulation results show that our Agent can increase the capacity for flow entries to 6x ~ 18x.

中文摘要 I
英文摘要 II
目錄 III
圖目錄 V
表目錄 VII
符號說明 VIII

第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 1
1.3 論文架構 2

第二章 文獻探討 3
2.1 軟體定義網路 3
2.2 OpenFlow 4
2.3 三態內容定址式記憶體 4
2.4 相關文獻 4

第三章 系統架構 6
3.1 異質性流量表整合（Heterogeneous Tables Integration）8
3.2 流量表分割與化簡（Table Division and Reduction）12
3.3 虛擬Packet In（Pseudo Packet In）16

第四章 模擬結果與分析 22
4.1 模擬方法 22
4.2 流量表分割與化簡（Table Division and Reduction）之效果分析 25
4.3 虛擬Packet In（Pseudo Packet In）之效果分析 29
4.4 門檻值TDRTH之探討 31
4.5 異質性流量表整合（Heterogeneous Tables Integration）之效果分析 32

第五章 結論與未來展望 35
5.1 結論 35
5.2 未來展望 36

參考文獻 37

附錄A 模擬數據 39
附錄B 發表之論文：Heterogeneous Flow Table Integration for Capacity Enhancement in Software-Defined Networks 45
附錄C 發表之論文：High-Efficiency Matching Mechanism for Off-Chip Tables in OpenFlow-enabled Legacy Switch 52

[1]. “Software-defined networking: The new norm for networks,” Palo Alto, CA, USA, White Paper, Apr. 2012. [Online]. Available: https://www.opennetworking.org/images/stories/downloads/whitepapers/ wp-sdnnewnorm.pdf
[2]. “Openflow Spec 1.3,” [Online]. Available: https://www.opennetworking.org/images/stories/down-loads/sdn-resources/onf-specifications/openflow/openflow-spec-v1.3.0.pdf
[3]. Y. Nakagawa, K. Hyoudou, C. Lee, S. Kobayashi, O. Shiraki, and T. Shimizu, “DomainFlow: Practical Flow Management Method Using Multiple Flow Tables in Commodity Switches,” in Proceedings of the Ninth ACM Conference on Emerging Networking Experiments and Technologies, Santa Barbara, California, USA, pp. 399-404, 2013.
[4]. H. Ma, Y. Yang, and Z. Mi, “A distributed storage framework of FlowTable in software defined network,” in Computers and Electrical Engineering, vol. 43, no. C, pp. 155-168, 2015.
[5]. N. Katta, O. Alipourfard, J. Rexford, and D. Walker, “CacheFlow: Dependency-Aware Rule-Caching for Software-Defined Networks,” in Proceedings of the Symposium on SDN Research, Santa Clara, CA, USA, pp. 1-12, 2016.
[6]. S. Kandula, S. Sengupta, A. Greenberg, P. Patel and R. Chaiken, “The Nature of Data Center Traffic: Measurements & Analysis,” in Proceedings of the 9th ACM SIGCOMM conference on Internet measurement, Chicago, Illinois, USA, pp. 202-208, 2009.
[7]. A. Shalimov, D. Zuikov, D. Zimarina, V. Pashkov and R. Smeliansky, “Advanced study of SDN/ OpenFlow controllers,” in Proceedings of the 9th Central & Eastern European Software Engineering Conference, Moscow, Russia, pp. 1-6, 2013.
[8]. W. Xia, Y. Wen, C. H. Foh, D. Niyato and H. Xie, “A Survey on Software-Defined Networking,” in IEEE Communications Surveys & Tutorials, vol. 17, no. 1, pp. 27-51, 2015.
[9]. V. Tanyingyong, M. Hidell and P. Sjödin, “Using hardware classification to improve PC-based OpenFlow switching,” in IEEE 12th International Conference on High Performance Switching and Routing, Cartagena, pp. 215-221, 2011.
[10]. H. Riaz, A. A. Bhatti, M. A. Tahir and M. Sarwar, “High speed content addressable memory with reduced size and less power consumption,” in International Conference on Design and Technology of Integrated Systems in Nanoscale Era, Istanbul, pp. 1-6, 2016.
[11]. Y. C. Cheng and P. C. Wang, “Scalable Multi-Match Packet Classification Using TCAM and SRAM,” in IEEE Transactions on Computers, vol. 65, no. 7, pp. 2257-2269, 2016.
[12]. C. R. Meiners, A. X. Liu, E. Torng and J. Patel, “Split: Optimizing Space, Power, and Throughput for TCAM-Based Classification,” in ACM/IEEE Seventh Symposium on Architectures for Networking and Communications Systems, Brooklyn, NY, pp. 200-210, 2011.
[13]. “Specification of Pica8 P5401”, [Online]. Available: http://www.stordis.com/fileadmin/user_up-load/vendors/Pica8/pica8_P-5401_32_x_40GbE_datasheet.pdf
[14]. “Specification of Cisco 93180-YC”, [Online]. Available: http://www.cisco.com/c/en/us/prod-ucts/collateral/switches/ nexus-9000- series-switches/datasheet-c78- 736651.html
[15]. H. Pan, H. Guan, J. Liu, W. Ding, C. Lin, and G. Xie, “The FlowAdapter: Enable Flexible Multi-table Processing on Legacy Hardware,” in Proceedings of the Second ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking, Hong Kong, China, pp. 85–90, 2013.
[16]. J. Ge, Z. Chen, Y. Wu, and Y. E, “H-SOFT: A Heuristic Storage space Optimisation Algorithm for Flow Table of OpenFlow,” in Concurrency and Computation: Practice and Experience, vol. 27, no. 13, pp. 3497–3509, 2015.
[17]. N. Matsumoto and M. Hayashi, “LightFlow: Speeding Up GPU-based Flow Switching and Facilitating Maintenance of Flow Table,” in IEEE 13th International Conference on High Performance Switching and Routing, Belgrade, pp. 76-81, 2012.
[18]. K. Qiu, Z. Chen, Y. Chen, J. Zhao and X. Wang, “GFlow: Towards GPU-based high-performance table matching in OpenFlow switches,” in International Conference on Information Networking, Cambodia, pp. 283-288, 2015.
[19]. M. Dong, H. Li, K. Ota, and J. Xiao, “Rule caching in SDN-enabled mobile access networks,” in IEEE Network, vol. 29, no. 4, pp. 40–45, 2015.
[20]. “Statistical Result of IPv6”, [Online]. Available: https://www.google.com/intl/en/ipv6/statis-tics.html#tab=ipv6-adoption&tab=ipv6-adoption
[21]. D. Lee, B. E. Carpenter and N. Brownlee, “Observations of UDP to TCP Ratio and Port Numbers,” in Fifth International Conference on Internet Monitoring and Protection, Barcelona, pp. 99-104, 2010.
[22]. Z. Chen, Y. Wu, J. Ge, and E. Yuepeng, “A New Lookup Model for Multiple Flow Tables of Open Flow with Implementation and Optimization Considerations,” in IEEE International Conference on Computer and Information Technology, pp. 528–532, 2014.
[23]. “Real Internet traffic”, [Online]. Available: http://tcpreplay.appneta.com/wiki/captures.html#bigflows -pcap
[24]. T. S. Chen, D. Y. Lee, T. T. Liu and A. Y. Wu, “Dynamic Reconfigurable Ternary Content Addressable Memory for OpenFlow-Compliant Low-Power Packet Processing,” in IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 63, no. 10, pp. 1661-1672, 2016.