隨著雲端運算的興起且由於雲端環境的高擴充性，各種雲端的應用服務也隨之增加。雲端應用服務品質會受到底層IaaS服務品質的影響，並且當雲端應用服務發生異常時，是否能即時辨別問題來源，會影響到問題的解決速度及應用服務的品質，因此雲端應用服務品質監控是一個很重要的議題。目前已有的雲端監控，主要以IaaS資源、安裝於其上之應用程式及SLA監控為主，若將現有的IaaS監控方式運用到雲端應用服務，僅能獲得服務或資源狀況，無法辨別服務問題來源。本研究以CSEP雲端安全實驗平台為例，轉換傳統品質控制流程步驟運用於雲端服務監控機制，並以定期監控方式發現服務問題、追蹤監控的方式辨別問題來源。 本研究雲端服務監控機制經實驗證明可有效偵測雲端應用服務CSEP平台共計六種主要服務之服務問題及雲端基礎設施供應商SAMEVED之API服務異常情形，主動發現服務異常問題且能提供錯誤訊息提示給使用者，並能即時辨別錯誤來源告知雲端服務管理者，幫助提升問題發現與解決速度，提供做為未來雲端應用服務進行監控設計時作為參考。 With the rise of Cloud Computing, cloud application services have become more popular and will be affected by IaaS’s quality of service. The problems cloud application services encountered be recognized or not will affect problem-solving time and quality of service, so quality control of cloud application service is an important issue. There are some researches and tools for IaaS resources, application and SLA monitoring. Using these cloud monitoring methods for cloud application services can only check the status of the service but not recognize the problems. We proposed a mechanism that used traditional quality control process for cloud services monitoring, and used CSEP Cloud Security Experimental Platform as example. Our mechanism including a period monitor and a trace monitor for finding problems. The cloud monitoring mechanism we proposed has been proven that it can recognize six CSEP main service problems and IaaS SAMEVED’s API service problems. Our mechanism can find the problems and provide some error messages to CSEP admin and users help for problem-solving and for giving other cloud application services monitoring as a reference.
