Through Virtual Machine Migration and Proactive Defense Resource Allocation to Maximize Network Survivability in a Cloud Environment

• Main Authors: Chia-Jung Chang, 張家榮
• Other Authors: Yeong-Sung Lin
• Format: Others
• Language: en_US
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/40132979089858212896

碩士 === 國立臺灣大學 === 資訊管理學研究所 === 103 === In recent years, the development of information technology not only brings much convenient to us but further changes our life style. The cloud computing is one of the information technology. Owing to the cloud computing, users can subscribe different kinds of flexible and scalable services on demand. This feature makes users spend less expense on establishing and maintaining IT infrastructure which attracts many enterprises to build their IT environment through the cloud platform. However, the information security issue is one of the must considered issue to enterprises who are going to adopt the cloud computing. In order to make the enterprise in the cloud environment not interrupt their abilities to provide services, VM Migration mechanism is a useful defense mechanism for avoiding service being disrupted. VM Migration mechanism enables VM dynamically move from one physical server to another and the mechanism can prevent VM from the malicious attack and further increase the network survivability. Besides, the importance of data for company has become more and more significant due to the rise of “Big Data Analysis”. It is special concerned by companies on how to prevent data being compromised by malicious attackers. Secret sharing mechanism is an effective mechanism to prevent data being stolen from attackers, it will cut data into N pieces and need K pieces to reconstruct the data for increasing the difficulty to steal data. Our goal is to help enterprises find out an effective way to prevent external malicious attacks to avoid service disruption or data stolen. In addition to traditional defense strategies such as firewall, IDS, IPS, we will take VM Migration mechanism and secret sharing mechanism introduced before to increase the network survivability. Beside, we will aid enterprises to seek the optimal way of allocating defense resource under limit budget to fight against the invasion of attackers. In our work, since many complex and full of randomness problem, we tried to take Mathematical Programming and Monte Carlo Simulation methods to solve this problem, and we look forward to helping defenders to improve the network survivability through effective allocation of resource.