臺灣博碩士論文加值系統

相較於傳統網路，網路功能虛擬化具有配置上的靈活性與擴展性，其中OpenStack提供良好的虛擬化技術與軟體定義網路提供集中管理資料平面的功能，運用這兩項技術使網路功能虛擬化快速的發展。在本篇論文中，透過OpenStack與軟體定義整合為網路服務鏈，網路服務鏈可以按照使用者網路需求動態新增或刪除網路功能虛擬機和調整傳輸路徑，為網路架構提高擴展性。網路服務鏈的高擴展性，提高網路服務鏈的使用者大量增加，也因此帶來虛擬機資源負荷量提升，造成網路服務虛擬機效能低落，所以需要一套負載平衡方法，幫助緩解虛擬機負載量。本篇論文提出的方法為在虛擬機設置資源門檻上限，連續時間內超過門檻限制則新增同功能虛擬機，並配合軟體定義網路轉移部份使用者至新虛擬機內服務，幫助緩解原虛擬機負荷量。本論文實作的結果可以快速建立網路服務功能虛擬機，並動態增加虛擬機，使用者封包路徑可以重新導向新的網路服務虛擬機，紓解原虛擬機負荷量，增強虛擬網路服務鏈的使用率。

Network function virtualization (NFV) compares with the traditional network, NFV has high flexibility and scalable, OpenStack which provides a good virtualization technology and Software-Defined Network (SDN) provides centralized management data plane features, Using OpenStack and SDN these technology enable network virtualization to grow quickly. In this thesis, OpenStack integrate with SDN to implement the network service chain (NSC). NSC can be dynamic add or delete network function VM and adjusting packet path based on user demand to improve scalability for network infrastructure. The high scalability of the NSC increase a lot of users, but also bringing VM resource overload and causing low efficiency of network ser-vices VM, so VM resource need a load balancing method. This thesis proposes a method to set the upper threshold of the virtual machine resources, if the traffic load in continuous time exceeds the threshold limit then OpenStack adds the VM of the same function and SDN transfer part of the user to the new VM to get the network service. The dynamic adding VM and redirecting path method can help mitigation the original virtual machine load. This thesis implements the experiment that can be quickly set up virtual network services with the OpenStack and SDN, dynamically add the same function VM then user packets can adjust the path to remove original VM highly overload state and enhanced the utilization of virtual network service chain.

1 Introduction 1
2 Related Works 4
2.1 Related works 4
2.2 Motivation 7
3 Preliminaries 9
3.1 System architecture 9
3.2 Basic idea 12
4 SDN-Based Network Service Chaining Platform Building 15
4.1 OpenStack and SDN 15
4.2 Building the SDN-based Network Service Chaining Platform Steps 20
5 Load Balancing Algorithm 23
5.1 VM Threshold Limit Algorithm 23
6 Implementation of the NSC Platform and Load Balancing analysis 29
6.1 Implementation of the SDN-based Network Service Chaining 29
6.2 Load Balancing Experiment Data Analysis 37
6.2.1 Experiment 1: To verify the CPU is decrease after using the load balancing mechanism 38
6.2.2 Experiment 2: To verify the stability is increase after using the load balancing mechanism 39
6.2.3 Experiment 3: To verify the request with multi transmission delay is increasing stability after using the load balancing mechanism 40
6.2.4 Experiment 4: To verify the throughput is increase after using the load balancing mechanism 41
6.2.5 Experiment 5: To verify the average throughput is increase after using the load balancing mechanism 42
7 Conclusions 43
8 Acknowledgements 44

[1] OpenStack Foundation, “OpenStack Installation Guide for Ubuntu 14.04,” http://docs.openstack.org/juno/install-guide/install/apt/content/ch_preface.html
[2] ETSI-ISG, “Network Functions Virtualisation - Introductory White Paper,” https://portal.etsi.org/nfv/nfv_white_paper.pdf
[3] ETSI-ISG, “Network Functions Virtualisation,” http://www.etsi.org/technologies-clusters/technologies/nfv
[4] ETSI-ISG, “Network Functions Virtualisation Use Cases,” http://www.etsi.org/deliver/etsi_gs/NFV/001_099/001/01.01.01_60/gs_NFV001v010101p.pdf
[5] IETF, “Service Function Chaining work group,” http://datatracker.ietf.org/wg/sfc
[6] Open Networking Foundation, “OpenFlow,” https://www.opennetworking.org/index.php
[7] Openstack: Open source cloud computing software, “Open source software for creating private and public clouds,” https://www.openstack.org/
[8] Open vSwitch, “Open vSwitch,” http://openvswitch.org/.
[9] J. Halpern and C. Pignataro, “Service Function Chaining architecture document,” https://tools.ietf.org/html/draft-ietf-sfc-architecture-0945
[10] ETSI-ISG, “Network Functions Virtualisation - White Paper3,” http://portal.etsi.org/NFV/NFV_White_Paper3.pdf
[11] W. John, K. Pentikousis, G. Agapiou, E. Jacob, M. Kind, A. Manzalini, F. Risso,D. Staessens, R. Steinert, and C. Meirosu, “Research Directions in Network Service
Chaining,” in Proceedings of IEEE SDN Future Networks and Services (SDN4FNS 2013),pp. 1–7, Trento, Italy, Nov. 2013.
[12] J. Blendin, J. Ruckert, N. Leymann, G. Schyguda, and D.Hausheer, “Position Paper:Software-Defined Network Service Chaining,” in Proceedings of IEEE Software Defined
Networks on European Workshops (EWSDN 2014), pp. 109–114, Budapest, Hungary, Sept. 2014.
[13] L. Chen, H. Shen, and K. Sapra, “RIAL: Resource Intensity Aware Load balancing in clouds,” in Proceedings of IEEE INFOCOM 2014, pp. 1294–1302, Toronto, Canada, May2014.
[14] A. Jain, A. Yadav, L. Namboodiri, and J. Abraham, “A Threshold Band Based Model for Automatic Load Balancing in Cloud Environment,” in Proceedings of IEEE International
Conference on Cloud Computing in Emerging Markets (CCEM 2013), pp. 1–7, Bangalore, Indian, Oct. 2013.
[15] S. Kaur, J. Singh, K. Kumar, and N. Ghumman, “Round-robin based load balancing in Software Defined Networking,” in Proceedings of International Conference on Comput-
ing for Sustainable Global Development (INDIACom 2015), pp. 2136–2139, New Delhi, Indian, March 2015.
[16] S. Domanal and G. Reddy, “Optimal load balancing in cloud computing by efficient utilization of virtual machines,” in Proceedings of International Conference on Communication Systems and Networks (COMSNETS 2014), pp. 1–4, Bangalore, Indian, Jan. 2014.
[17] T. Janpan, V. Visoottiviseth, and R. Takano, “A virtual machine consolidation framework for CloudStack platforms,” in Proceedings of International Conference on Information
Networking (ICOIN 2014), pp. 28–33, Phuket, Thailand, Feb. 2014.
[18] NTT Communications, “RYU Controller,” http://osrg.github.io/ryu/.
[19] “VLAN,” https://en.wikipedia.org/wiki/Virtual_LAN.
[20] “GRE,” https://en.wikipedia.org/wiki/Generic_Routing_Encapsulation.
[21] “VxLAN,” https://en.wikipedia.org/wiki/Virtual_Extensible_LAN.
[22] “REST,” https://zh.wikipedia.org/wiki/REST.
[23] Gary Chen and AI Gillen, “KVM for Server Virtualization: An Open Source Solution Comes of Age - IBM white paper,” http://www03.ibm.com/systems/resources/systems_virtualization_idc_kvmforservervirtualization.pdf
[24] “Hypervisor,” https://en.wikipedia.org/wiki/Hypervisor.
[25] “Jemeter,” http://jmeter.apache.org/usermanual/glossary.html.