- 學位論文
雲端服務之系統性能與資安事件的監控
The Monitoring of System Performance and Security Event for Cloud Services
摘要
最近幾年，雲端運算(Cloud Computing)已成為資訊產業中資訊系統彈性佈署及功能分配之受矚目的新興議題。雲端服務(Cloud Services)採用開放式的架構，企業將公共服務之資訊儲存在雲端，透過大型資訊平台佈署即時提供全球性的無所不在(ubiquitous)的服務方式，是目前企業邁向服務差異化的重要資訊科技之一，但服務管理與資安防護的問題一直困擾著想採用雲端服務的主管。根據市場調查，企業未採用雲端服務，其中一項因素是因為雲端服務的管理上有著的安全考量之操作之複雜性。若能增進雲端服務管理之跨系統便利性與嚴密的資安監控，將可提升使用者之資訊服務的品質(QoS)。本研究參考代理人(Agent)運作模式，建立一套系統性能監控回報系統，執行雲端服務即時狀況之資訊蒐集及回報，當出現服務異常或安全警訊，監控回報系統可透過簡訊快速反應給遠方的管理人員，透過遠端立即處理及警示事件的統計分析，可有效降低管理者之負擔並強化雲端服務使用者的信心。
並列摘要
Nowadays, Cloud Computing has already gained attention of Information Technology (IT) industry for flexible deployment of information system and function distribution of services. Cloud Services provides the functionality to enterprises for shifting their services information to the cloud, and provide ubiquitous services globally by deploying through massive information platform on an open structure of information platform. Obviously, it is one of the most important IT strategies for enterprises that move toward service differential, but problems of service management and information security are still bothering managers of information management. According to the industrial survey, the one of the reasons why enterprises do not adopt cloud services solutions is because of concerning security issue and managing a systems operational complexity. If one can improve the operational compatibility of cross-platform and offer a more secure matter of information monitoring for cloud service management system, then it could massively improve the Quality of Service (QoS) for service consumer. This paper constructs a system performance monitoring and report platform which collects and retrieves information from on-line cloud services using agent technology. When a service reports any malfunction or security incident, our platform will immediately sends alert messages to administrators such that could effectively lower administrators' burden and put the user faith for cloud services by remotely handling the alert events and analyzing incident-related data.