臺灣博碩士論文加值系統

近幾年來，行動寬頻的網路傳輸量不斷地上升，大部分的流量主要都是來自於高互動的應用服務，如虛擬實境、工業物聯網及機器類型通訊(Machine-Type Communication)，低延遲的需求將被視為第五代網路通訊標準制定的首要關鍵任務。為了滿足低延遲的需求，行動邊緣運算的概念被提了出來，將雲端運算的計算資源下放至邊緣網路，以及透過虛擬化的技術，讓服務提供商租用計算資源，或者網路營運商部署其虛擬網路功能(VNF)至邊緣網路，降低網路的延遲，然而如何適當地將服務功能部署到邊緣網路將會是問題，部署的結果將會影響到整體使用者的效能。
本論文所提出的GASDE是一種高效能的部署策略，用於邊緣網路環境中的服務功能部署。GASDE使用了基因演算法，在考慮多個租戶租用邊緣運算的計算資源之情況下，降低用戶端存取服務的平均網路延遲，並且在部署決策時考慮了服務功能部署的成本。GASDE部署策略不僅能用在純邊緣運算情境的部署，還能用在同時考慮邊緣運算以及雲端運算情境的部署。模擬結果顯示，與其他2種部署策略 : GRE以及DCB相比，無論是在純邊緣運算的情境或是同時考慮邊緣運算以及雲端運算的情境，在網路延遲和服務功能部署成本的表現上，均表現出較佳的效能。此外，本論文還在XenServer中設計並實作了一個服務功能邊緣平台，驗證邊緣運算對於網路延遲的重要性，以及本論文所提出的演算法之可應用性。

In recent years, the mobile data traffic has a tremendous growth, especially most of these traffic originate from highly interactive applications such as virtual reality, Internet of Things (IoT) and Machine-Type Communication(MTC). The demand for low-latency communications has been considered as one of critical issue for fifth-generation standardization. In order to satisfy the demand of low-latency, the concept of mobile edge computing is recently emerged by placing computation resource to the edge network. With the technology of virtualization, service providers can rent computation resource from the infrastructure of network operator, and network operators also can deploy service functions(SFs) to the edge network to reduce the network latency. However, how to appropriately deploy these service functions into edge network will be a problem.
We propose GASDE, a high-performance approach for deploying service functions into the edge network. GASDE uses genetic-algorithm(GA) to reduce network delay and cost of deployment, which considers the situation multi-tenancy would deploy their service functions into edge network. The result of simulation shows that when compared with other two strategies: GRE and DCB has the better performance of network delay and cost of deployment no matter in considering the case of only edge computing or cloud edge computing. We also implement a service function edge platform in XenServer to verify our works are more comprehensive and realistic.

第一章 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
第二章 背景知識與相關研究 5
2.1 邊緣運算 5
2.1.1 從雲端運算到邊緣運算 5
2.1.2 邊緣運算(Edge Computing) 6
2.1.3 Multi-access Edge Computing 8
2.1.4 多租戶(Multi-Tenancy) 11
2.2 網路功能虛擬化 12
2.3 基因演算法 14
2.4 虛擬網路功能之服務部署 17
2.5 相關研究比較 21
第三章 研究方法 23
3.1 系統架構與設計 23
3.1.1 Hypervisor 25
3.1.2 OVS TC Agent 26
3.1.3 Computation Task Server 26
3.1.4 OpenFlow Control 27
3.1.5 Tunnel Module 28
3.1.6 Health Care 28
3.1.7 Request Redirection 29
3.1.8 SF Placement Simulator 30
3.1.9 Flow Rule Production 31
3.1.10 Packet Handler 32
3.2系統運作流程與機制 32
3.2.1 系統假設與定義 33
3.2.2 資料符號表 35
3.2.3 GASDE運作流程 39
3.3 系統實作 48
第四章 實驗與討論 52
4.1 情境一 : 服務功能邊緣平台測試 52
4.1.1實驗一 : 邊緣運算對於雲端運算之效能影響 52
4.1.2實驗二 : 服務功能壓力測試以即時串流為例 59
4.1.3實驗三 : 隧道技術頻寬效能分析 62
4.1.4實驗四 : 實際部署之反應時間測試與比較 65
4.2 情境二 GASDE於邊緣運算環境部署之成本及延遲 68
4.2.1 實驗五 : Small-scale網路拓撲下部署的網路延遲比較 70
4.2.2 實驗六 : Small-scale網路拓撲下部署的部署成本比較 75
4.2.3 實驗七 : Large-scale網路拓撲下部署的網路延遲比較 77
4.2.4 實驗八 : Large-scale網路拓撲下部署的部署成本比較 80
4.3 情境三 : GASDE於同時考慮邊緣運算以及雲端運算環境部署之成本及延遲 82
4.3.1 實驗九 : 當計算資源需求低的服務功能Request數目較多之部署分析 83
4.3.2 實驗十 : 當各服務功能Request數目皆相同時之部署分析 86
第五章 結論與未來研究方向 88
5.1 結論 88
5.2未來研究 89
參考文獻 91

[1] Wikipedia, "Software-defined networking", 2018. [2018]. Available: https://en.wikipedia.org/wiki/Software-defined_networking
[2] 'NFV - Network Functions Virtualization'. Available: http://vega-bi.blogspot.tw/2014/08/nfv-network-functions-virtualization.html
[3] Wikipedia, "Edge Computing", 2018. [Online]. Available: https://en.wikipedia.org/wiki/Edge_computing
[4] Cisco Visual Networking Index : Global Mobile Data Traffic Forecast, 2015-2020. Available: https://www.cisco.com/c/en/us/solutions/service-provider/visual-networking-index-vni/index.html
[5] etsi.org, "Multi-access Edge Computing", 2018. [Online]. Available: http://www.etsi.org/technologies-clusters/technologies/multi-access-edge-computing
[6] K. Samdanis, X. Costa-Perez, and V. Sciancalepore, "From network sharing to multi-tenancy: The 5G network slice broker," IEEE Communications Magazine, vol. 54, no. 7, pp. 32-39, 2016.
[7] vCPE (virtual customer premises equipment), 2018. [Online]. Available: https://searchsdn.techtarget.com/definition/vCPE-virtual-customer-premise-equipment
[8] H. Tian, J. Wu, and H. Shen, "Efficient Algorithms for VM Placement in Cloud Data Centers," in 18th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT), 2017, pp. 75-80.
[9] X. Meng, V. Pappas, and L. Zhang, "Improving the Scalability of Data Center Networks with Traffic-aware Virtual Machine Placement," in Proceedings IEEE INFOCOM, 2010, pp. 1-9.
[10] T. Yapicioglu and S. Oktug, "A Traffic-Aware Virtual Machine Placement Method for Cloud Data Centers," in IEEE/ACM 6th International Conference on Utility and Cloud Computing, 2013, pp. 299-301.
[11] O. Biran et al., "A Stable Network-Aware VM Placement for Cloud Systems," in 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (ccgrid 2012), 2012, pp. 498-506.
[12] D. Cho, J. Taheri, A. Y. Zomaya, and L. Wang, "Virtual Network Function Placement: Towards Minimizing Network Latency and Lead Time," in IEEE International Conference on Cloud Computing Technology and Science (CloudCom), 2017, pp. 90-97.
[13] M. Xia, M. Shirazipour, Y. Zhang, H. Green, and A. Takacs, "Network Function Placement for NFV Chaining in Packet/Optical Datacenters," Journal of Lightwave Technology, vol. 33, no. 8, pp. 1565-1570, 2015/04/15.
[14] K. Phemius and M. Bouet, "Monitoring latency with openflow," in Network and Service Management (CNSM), 9th International Conference on, 2013, pp. 122-125: IEEE.
[15] J. Xu, B. Palanisamy, H. Ludwig, and Q. Wang, "Zenith: Utility-Aware Resource Allocation for Edge Computing," in IEEE International Conference on Edge Computing (EDGE), 2017, pp. 47-54.
[16] L. Zhao, J. Liu, Y. Shi, W. Sun, and H. Guo, "Optimal Placement of Virtual Machines in Mobile Edge Computing," in GLOBECOM IEEE Global Communications Conference, 2017, pp. 1-6: IEEE.
[17] S. Wang, R. Urgaonkar, M. Zafer, T. He, K. Chan, and K. K. Leung, "Dynamic service migration in mobile edge-clouds," in IFIP Networking Conference (IFIP Networking), 2015, pp. 1-9.
[18] R. Cziva and D. P. Pezaros, "On the Latency Benefits of Edge NFV," in ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS), 2017, pp. 105-106.
[19] T. Taleb, K. Samdanis, B. Mada, H. Flinck, S. Dutta, and D. Sabella, "On Multi-Access Edge Computing: A Survey of the Emerging 5G Network Edge Cloud Architecture and Orchestration," IEEE Communications Surveys & Tutorials, vol. 19, no. 3, pp. 1657-1681, 2017.
[20] IIoT Edge Computing vs. Cloud Computing, 2018. [Online]. Available: https://openautomationsoftware.com/blog/iiot-edge-computing-vs-cloud-computing/
[21] MEC Deployments in 4G and Evolution Towards 5G. Available: http://www.etsi.org/images/files/ETSIWhitePapers/etsi_wp24_MEC_deployment_in_4G_5G_FINAL.pdf
[22] etsi.org, 'Network Functions Virtualisation (NFV) Architectural Framework', 2018. [Online]. Available: http://www.etsi.org/deliver/etsi_gs/NFV/001_099/002/01.02.01_60/gs_NFV002v010201p.pdf
[23] Xenserver.org,"XenServer | Open Source Server Virtualization", 2017. [Online]. Available: https://xenserver.org/
[24] OpenStack.org, 'Open source software for creating private and public clouds', 2018. [Online]. Available: https://www.openstack.org/
[25] Wikipedia, "Genetic algorithm", 2018. [Online]. Available: https://en.wikipedia.org/wiki/Genetic_algorithm
[26] The Genetic Algorithm - Explained, 2018. [Online]. Available: http://techeffigytutorials.blogspot.com/2015/02/the-genetic-algorithm-explained.html
[27] OpenStack.org, 'Tacker - OpenStack NFV Orchestration', 2017. [Online]. Available: https://wiki.openstack.org/wiki/Tacker
[28] Kubernetes, 2018. [Online]. Available: https://kubernetes.io/
[29] Q. Fan and N. Ansari, "Cost Aware cloudlet Placement for big data processing at the edge," in IEEE International Conference on Communications (ICC), 2017, pp. 1-6.
[30] 林雋策 and C.-T. Lin, "基於SDN與NFV的資源調度器應用於虛擬網路功能部署及高可用性 - 以即時影音串流服務為例;SDN/NFV-based Resource Orchestrator for VNF Deployment and High Availability - A Case Study of Live Streaming Service," 國立中央大學.
[31] X. Sun and N. Ansari, "Latency Aware Workload Offloading in the Cloudlet Network," IEEE Communications Letters, vol. 21, no. 7, pp. 1481-1484, 2017.
[32] archive.openflow.org. 'OpenFlow: Enabling Innovation in Campus Networks', 2018. Available: http://archive.openflow.org/documents/openflow-wp-latest.pdf
[33] Openvswitch.org, 'Production Quality, Multilayer Open Virtual Switch', 2018. Available: http://openvswitch.org/
[34] tc(8) - Linux man page. Available: https://linux.die.net/man/8/tc
[35] Squeaky-pl. "japronto". Available: https://github.com/squeaky-pl/japronto
[36] Ryu SDN Framework, 2018. [Online]. Available: https://osrg.github.io/ryu/
[37] Internet Engineering Task Force(IETF)."Generic Routing Encapsulation(GRE)"RFC1701. Available: https://tools.ietf.org/html/rfc1701
[38] "Internet Engineer Task Force(IETF)."Virtual eXtensible Local Area Network (VXLAN): A Framework for Overlaying Virtualized Layer 2 Networks over Layer 3 Networks"RFC7348."
[39] Wikipedia, "Floyd-Warshall algorithm". Available: https://en.wikipedia.org/wiki/Floyd%E2%80%93Warshall_algorithm
[40] Y. Mao, C. You, J. Zhang, K. Huang, and K. B. Letaief, "A Survey on Mobile Edge Computing: The Communication Perspective," IEEE Communications Surveys & Tutorials, vol. 19, no. 4, pp. 2322-2358, 2017.
[41] btfak. "sniper". Available: https://github.com/btfak/sniper
[42] ffmpeg.org, ' ffmpeg ' , 2018. [Online]. Available: https://ffmpeg.org/