臺灣博碩士論文加值系統

從網絡運營商的角度來看，網絡管理變得越來越困難，因為網絡中的每個組件都按照自己的演算法處理數據包。軟件定義網絡（SDN）把數據平面和控制平面分離，並使用軟件程序來控制整個網絡的行為，為網絡管理提供了新的可能性。憑藉其技術，管理大型的數據中心會更容易，因為網絡中的每個組件都根據控制器的指令處理數據包。但龐大的網絡結構會令網絡運營商難以在第一時間發現網絡中出現的問題，大多是依靠用戶回報。當網絡遭受一些惡意攻擊時，網絡運營商如果無法立即找到並解決該漏洞，將會蒙受重大損失。
在本文中，我們為執行網絡診斷和故障排除提供一個有更效的方法。我們的技術使得網絡運營商能夠通過用已訓練的機器學習模型分析一系列有時間性的網路拓撲來了解當前所控制的網絡狀態。此模型藉著調整訓練方向，可用於探查網絡上所運行的服務、檢測網絡是否出現異常狀態，從而發展出偵錯、備份、還原等多方面的應用。

From the perspective of network operators, network management is becoming much more difficult since every component in the network processes packets according to its own algorithm. A paradigm in networking, software defined network (SDN), advocates separating the data plane and the control plane and uses a logically centralized software program to control the behavior of the entire network. SDN introduces new possibilities for network management and configuration methods. With its technology, it is easier to manage a large data center since every component in the network processes packets according to the control plane’s instructions. It is difficult for network operators to find the problems in a large network at the first timing. They rely on the users' responding for the most part. When the network subjected to some malicious attacks, the network operator would suffer a significant loss if they cannot find and respond to the issue instantly.
In this article, we focus on providing better visibility for performing network diagnosis and troubleshooting. The technologies we describe enable network operators to know the state of the controlling network by analyzing a series of time-related network views with a trained machine learning model. With the adjustment of training direction, the model can be used to probe the service running on network or detect network abnormal state. Thus developing the debugging, backup, restore and many other functions.

Chapter 1. Introduction 1
Chapter 2. Background 3
2.1. SDN and OpenFlow 3
2.2. High Availability 4
2.3. Related Work 4
Chapter 3. System Description 6
3.1. Network View 7
3.1.1. Dynamic Network View 9
3.1.2. Simple Graph and Difference Graph 10
3.1.3. Feature Extraction of Graphs 14
3.2. Application Oriented Network State Extraction and Modeling 16
3.2.1. Application Detector 16
3.2.2. Service Suspension Detector 17
3.2.3. API Template Detector 17
3.3. API Integrator and Northbound API Generation 19
Chapter 4. Experiments 26
4.1. Experiment 1 – Effect of Training Data Size on Modeling Accuracy (Multiple Switches) 27
4.2. Experiment 2 – Effect of Training Data Size on Modeling Accuracy (Single Switch) 28
4.3. Experiment 3 – Compatibility of Detectors (From Multiple to Single) 29
4.4. Experiment 4 – Compatibility of Detectors (From Single to Multiple) 30
Chapter 5. Conclusion 34
Reference… 35

[1] SDN. Available: https://www.opennetworking.org/sdn-resources/sdn-definition
[2] Floodlight Project. Available: http://www.projectfloodlight.org/
[3] Ryu Project. Available: https://osrg.github.io/ryu/
[4] N. McKeown et al., ”OpenFlow: Enabling Innovation in Campus Networks,” ACM Comp. Commun. Rev., Apr. 2008.
[5] Kim, Hyojoon, and Nick Feamster. "Improving network management with software defined networking." IEEE Communications Magazine 51.2 (2013): 114-119.
[6] Chowdhury, Shihabur Rahman, et al. "Payless: A low cost network monitoring framework for software defined networks." Network Operations and Management Symposium (NOMS), 2014 IEEE. IEEE, 2014.
[7] Khurshid, Ahmed, et al. "Veriflow: Verifying network-wide invariants in real time." ACM SIGCOMM Computer Communication Review 42.4 (2012): 467-472.
[8] Lee, Seungsoo, et al. "DELTA: A security assessment framework for software-defined networks." Proceedings of NDSS. 2017.
[9] Kazemian, Peyman, et al. "Real Time Network Policy Checking Using Header Space Analysis." NSDI. 2013.
[10] Kazemian, Peyman, George Varghese, and Nick McKeown. "Header Space Analysis: Static Checking for Networks." NSDI. 2012.
[11] Dhawan, Mohan, et al. "SPHINX: Detecting Security Attacks in Software-Defined Networks." NDSS. 2015.
[12] Henderson, Keith, et al. "It's who you know: graph mining using recursive structural features." Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2011.
[13] OpenFlow Specification. Available: https://www.opennetworking.org/images/stories/downloads/sdn-resources/onf-specifications/openflow/openflow-spec-v1.4.0.pdf
[14] Cortes, Corinna, and Vladimir Vapnik. "Support-vector networks." Machine learning 20.3 (1995): 273-297.
[15] Mininet. Available: http://mininet.org
[16] LIBSVM. Available: https://www.csie.ntu.edu.tw/~cjlin/