簡易檢索 / 詳目顯示

研究生: 何信昱
Hsin-Yu Ho
論文名稱: 5G雲端服務中虛擬機干擾感知的最佳化部署
The Optimal Deployment of Virtual Machine- Aware for 5G Cloud Computing Services
指導教授: 黎碧煌
Bih-Hwang Lee
口試委員: 黎碧煌
Bih-Hwang Lee
陳俊良
Jiann-Liang Chen
鍾添曜
Tein-Yaw Chung
馬奕葳
Yi-Wei Ma
學位類別: 碩士
Master
系所名稱: 電資學院 - 電機工程系
Department of Electrical Engineering
論文出版年: 2021
畢業學年度: 109
語文別: 英文
論文頁數: 60
中文關鍵詞: 5G服務雲端運算邊緣運算最佳化部署虛擬機干擾
外文關鍵詞: Service deployment, Deployment optimization, VM interference
相關次數: 點閱:150下載:1
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
  • 隨著5G行動網路的到來,以5G網路為基礎的雲端服務數量也隨之產生,例如:車聯網、4K/8K影片串流、智能電網等等,然而以現有的集中式雲端運算架構無法有效地滿足5G服務的延遲需求,因此邊緣運算架構的出現,使服務提供商能將服務移到邊緣雲就近部署,如此一來使用者得以更快地存取到5G服務,除此之外,對雲端運營商而言,邊緣運算也使得網路架構的設計變得更有彈性。
    雖然邊緣運算能有效降低存取服務的響應時間,但隨意地將服務盡量往邊緣雲部署也不是一個好的策略,不僅浪費了珍貴的邊緣雲資源,也造成嚴重的虛擬機干擾問題,進而導致部署在該伺服器上的網路服務效能下降。
    為了使部署的雲端服務效能不受虛擬機干擾影響,雲端服務商必須擁有一套能夠預估虛擬機干擾的影響程度並計算出最佳化部署的系統,因此本研究提出一套部署系統架構來承接使用者的服務請求,在最佳化部署方面,本研究提出了一套評估框架來計算服務部署對系統帶來的影響,並在感知虛擬機干擾的前提下,提出了一套以成本-效能分析找出最佳化部署的演算法。


    As the advent of 5G network, the number of 5G cloud computing services grew explosively, such as vehicle-to-everything, 4K/8K video streaming and smart grid. To fully fulfill the requirement of 5G services, the appearance of edge computing enable cloud operators to move services into edge cloud which is capable of reducing response time of accessing services.
    Even though edge computing is able to make 5G services available, deploying services into edge cloud as much as possible isn’t a good method, either. It not only wastes expensive edge cloud resources but also cause severe virtual machine interference on edge servers.
    To prevent deployed services from VM interference, cloud operators must have a system for tactically deploying services on limited servers. This system should have an ability to predict VM interference on a server and analyze for the optimal deployment. Hence, first of all, this study proposed a deployment system for handling service requests from end-users. Second, this study designed a framework of evaluating service deployments with different deploying algorithms. Last, this study proposed a Cost-Performance Analysis algorithm to derive the optimal deployments. From the simulation result, it is manifest that the proposed Cost-Performance Analysis could effectively satisfy the requirements of service requests and maximize the application performance of deployed services.

    摘要 Abstract Acknowledgments Table of Contents List of Abbreviations List of Notations List of Figures List of Tables Chapter 1 Introduction 1.1 Study Motivation 1.2 Contribution 1.3 Organization of this Thesis Chapter 2 Background Knowledge and Related Work 2.1 Overview of 5G scenarios uRLLC eMBB mMTC 2.2 Overview of Cloud Computing 2.3 Overview of Edge Computing Architecture 2.4 Virtual Machine Interference 2.5 Min-Max Normalization 2.6 Related Work  Adaptive Interference-Aware VNF Placement for Service-Customized 5G Network Slices A VNF Placement Method based on VNF Characteristics Physical and Virtual Resources Inventory Modeling for Efficient VNF Placement Towards Edge Slicing: VNF Placement Algorithms for a Dynamic & Realistic Edge Cloud Environment Coalitional Game for the Creation of Efficient Virtual Core Network Slices in 5G Mobile Systems Chapter 3 System Model 3.1 Problem Description 3.2 System Architecture 3.2.1 Service catalogue 3.2.2 VM controller 19 3.2.3 Optimization module 3.3 System Scenario 21 3.4 A Framework of Evaluating Service Deployments 3.4.1 Characteristics of a Server 3.4.2 Evaluation of VM Interference, System Performance and Application Performance 3.4.3 Evaluation of E2E Latency 3.5 Deployment Optimization 3.5.1 Optimization Objective 3.5.2 Evaluation module 3.5.3 Cost-Performance Analysis module Chapter 4 Performance Evaluation 4.1 Simulation Environment and Parameters 4.2 Compared Deploying Algorithms 4.3 Simulation Result 4.3.1 Average Normalized Application Performance 4.3.2 Percentage of Satisfied Services to All Service Requests 4.3.3 Average uRLLC E2E Delay 4.3.4 Average eMBB E2E Delay Chapter 5 Conclusion Chapter 6 Future Work Reference

    [1] S. Gyawali, S. Xu, Y. Qian and R. Q. Hu, "Challenges and Solutions for Cellular Based V2X Communications", Proceeding of the IEEE Communications Surveys & Tutorials, vol. 23, no. 1, pp. 222-255, 2021.
    [2] 5G E2E Technology to Support Verticals URLLC Requirements. https://www.ngmn.org/publications/5g-e2e-technology-to-support-verticals-urllc-requirements.html. Accessed on: 2021/6/29.
    [3] NGMN 5G WHITE PAPER. https://www.ngmn.org/work-programme/5g-white-paper.html. Accessed on: 2021/6/29.
    [4] Q. Zhang, F. Liu and C. Zeng, "Adaptive Interference-Aware VNF Placement for Service-Customized 5G Network Slices", Proceeding of the IEEE INFOCOM 2019 - IEEE Conference on Computer Communications, pp. 2449-2457, 2019
    [5] Aruna, K., Pradeep, G, “Performance and Scalability Improvement Using IoT-Based Edge Computing Container Technologies”, Proceeding of the SN COMPUT, 2020.
    [6] Y. Koh, R. Knauerhase, P. Brett, M. Bowman, Z. Wen and C. Pu, "An Analysis of Performance Interference Effects in Virtual Environments", Proceeding of the 2007 IEEE International Symposium on Performance Analysis of Systems & Software, pp. 200-209, 2007.
    [7] F. Xu, F. Liu, L. Liu, H. Jin, B. Li and B. Li, "iAware: Making Live Migration of Virtual Machines Interference-Aware in the Cloud," Proceeding of the IEEE Transactions on Computers, vol. 63, no. 12, pp. 3012-3025, 2014.
    [8] New Services & Application With 5G Ultra-Reliable Low Latency Communications. https://www.5gamericas.org/new-services-applications-with-5g-ultra-reliable-low-latency-communications. Accessed on: 2021/6/22.
    [9] A. Laghrissi, T. Taleb, M. Bagaa and H. Flinck, "Towards Edge Slicing: VNF Placement Algorithms for a Dynamic & Realistic Edge Cloud Environment,", Proceeding of the 2017 IEEE Global Communications Conference, pp. 1-6, 2017.
    [10] M. Bagaa, T. Taleb, A. Laghrissi, A. Ksentini and H. Flinck, "Coalitional Game for the Creation of Efficient Virtual Core Network Slices in 5G Mobile Systems", Proceeding of the IEEE Journal on Selected Areas in Communications, vol. 36, no. 3, pp. 469-484, 2018
    [11] S. I. Kim and H. S. Kim, "A VNF Placement Method based on VNF Characteristics", Proceeding of the 2021 International Conference on Information Networking (ICOIN), pp. 864-869, 2021
    [12] L. Popokh, P. Olive, I. Aldama, Y. Al-Doori and S. Nair, "Physical and Virtual Resources Inventory Modeling for Efficient VNF Placement", Proceeding of the 2020 IEEE 10th International Conference on Consumer Electronics (ICCE-Berlin), pp. 1-6, 2020

    QR CODE