Design and implementation of a portable and scalable fault tolerance mechanism in cloud environments

碩士 === 靜宜大學 === 資訊工程學系 === 99 === The growth of Cloud computing Technologies is found in impressive pace; more and more applications are available online. All these changes are still found in early stage of this development, and many new issues and topics are waiting for exploration. In this propose...

Full description

Bibliographic Details
Main Authors: Tu, Hua Yan, 涂化衍
Other Authors: Li, Kuan-Ching
Format: Others
Language:zh-TW
Published: 2011
Online Access:http://ndltd.ncl.edu.tw/handle/69823680812732940577
Description
Summary:碩士 === 靜宜大學 === 資訊工程學系 === 99 === The growth of Cloud computing Technologies is found in impressive pace; more and more applications are available online. All these changes are still found in early stage of this development, and many new issues and topics are waiting for exploration. In this proposed study, we have focused on the design and construction of a FT (Fault Tolerance) system in a private cloud environment. The main objective is to improve the QoS (Quality of service) in the cloud to provide more reliability and availability for users, and the main challenge of this research is the way to provide safer and stable VMs. In the proposed experimental environment, the testing design is performed on online VMs and keeping them online and working, even if an unknown problem suddenly cause system crash.  In order to solve unpredicted failures in a Cloud environment, we develop a GUI (Graphical User Interface) FT tool for a cloud system platform administrator, by integrating some most common and general open-source applications into this proposed tool, and disposing this on a server to process FT needs. There are three features in this FT tool, which are also cross-platform, time saving, and easy to use. These abilities are helpful to maintain a cloud system platform stable and operable. With this tool, users have additional protection against system crashing, losing important files and keeping all the settings on virtual OS. Experimental testings using this tool shown its efficiency and portability, and therefore, users encounter higher QoS for all services and applications in a given Cloud system platform.