Abstract: | 企業雲端運算(Enterprise Cloud Computing)是IT 服務產業的新趨勢,因為可以降低成本並提供服務的品質(QoS)。然而,把傳統企業運算服務轉移至雲端環境有很多的阻礙,比如如何確保服務的品質,包含安全性、可用性、資料隱私等等。從多重租約、跨層服務組合(SaaS、PaaS、IaaS)、使用者需求和服務層級協議(SLAs)的多重限制等問題讓等讓企業雲端運算(ECC)議題變得十分複雜。為了克服企業雲端運算的困難,本整合性計畫提出一套技術去確保服務的品質和IaaS、PaaS、SaaS 方面的運算成效。由於企業雲端運算的高複雜性,本計畫將建構包含SaaS、PaaS、IaaS 與服務的正規化塑模方法,以確保供應商所提供的服務品質協議可以滿足多重品質保證因素,例如安全性、效能、時效性、傳輸速度和可靠性,可同時對大量客戶需要的運算各種需求做系統化分析。透過正規化模組,系統中的的自動化分析模組與測試,可以達成自動化測試以及監控執行階段系統是否符合需求與品質服務協議等相關限制要求。為了使企業從傳統商業架構的 IT 技術移植到雲端計算環境下可以無風險之憂,我們所提出的整合型計畫技術為眼下企業所急需的。本計劃架構簡述如下,子計畫一 (IaaS):實作一個對具確保服務品質(QoSA)的企業雲端運算的基礎架構服務環境,這個雲端IaaS 環境由虛擬機器的分配和管理之資源負載平衡和節能功能組成。子計畫二 (PaaS):提供對企業雲端運算的服務組合之中介軟體和受QoS 限制與成效的企業雲端運算的監控機制和API。子計畫三(PaaS):提出eQDA:一個彈性的資料服務品質確保環境,去支援資料服務塑模,SLA 驗證和監控,以及確保資料服務品質。子計畫四(SaaS): 設計一套支援雲端服務和系統軟體重整工程至雲端環境的工具集,並建立元件相關需求與限制條件,以支援服務需求與品質服務協議。子計畫五(SaaS 行動商業):應用設計模式技術,服務導向架構去發展一套行動應用框架和中介軟體。子計畫六(SaaS):目標在解決基於建構雲端軟體自動化測試服務(TaaS)的議題,包括傳統程式測試,雲端環境程式測試和綠能雲端環境測試,以確保基於雲端程式的品質。本計畫的第一年將對環境塑模,包括佈署模組,服務組合模組和測試模組,故可以藉由這些模組分析來確保QoS 和SLA,這些方法和原型的工具將被設計和開發。第二年目標主要整合各子計畫和將本計畫套用至工業案例,如移植ERP,CRM 系統至雲端環境。第三年將應用本計畫所提之解決方法至實際軟體工業案例來證明本計畫之可行性,本計畫的開發將建構在開放原始碼之基礎上。 Enterprise Cloud Computing (ECC) is a new trend for IT service provisioning due to cost down and increase of Quality of Service (QoS). However, there are many obstacles to hinder the services moving from traditional enterprise computing to cloud environment, such as how to assure the quality of services, including security, availability, data privacy…etc. The problems from the issues of the multi-tenancy, cross layer service composition (SaaS, PaaS, IaaS), multiple constraints from user requirements and service level agreements (SLAs) make ECC very complex. In order to overcome the difficulties of ECC, this integrated project proposes technologies to assure the quality of service and effectiveness of ECC from IaaS, PaaS, and SaaS aspects. Due to the high complexity of ECC, a formal model of SaaS, PaaS, IaaS and services will be designed and constructed, so the analysis on the assurance of the service providers can satisfy multiple QoS factors, such as security, performance, timeliness, throughput, and reliability, for multiple consumers simultaneously and dynamically can be systematically analyzed. Based on the formal model, analysis model and testing model can be automatically generated to support automatic testing and as well as run time monitoring on requirements/SLAs constraints. In order to support the risk free migration of IT provisioning of enterprises from traditional enterprise architecture to cloud computing, the technologies proposed in this integrated project are urgently needed. Sub-project 1(IaaS) designs and implements a IaaS environment for ECC with quality of service assurance (QoSA). This cloud IaaS environment consists of resource load balancing and power saving functions for virtual machines allocation and management. Sub-project 2(PaaS) provides middle ware for the service composition of ECC as well as the monitoring mechanism and APIs for QoS enforcement and effectiveness of ECC. Sub-project 3(PaaS) proposes an elastic data service quality assurance environment eDQA, to support data service modeling, SLA verification & monitoring, and data service quality assurance. Sub-project 4(SaaS) designs supporting toolset for cloud service and a systematic reengineering methodology reengineering traditional software to cloud. Sub-project 5(SaaS: Mobile enterprise) applies the technologies of design patterns, service-oriented architecture to develop a mobile application framework and middleware. Sub-project 6(SaaS) aims to give solutions to the issues by constructing cloud-based software automatic testing services TaaS(Test as a Service), including traditional program testing, cloud-based program testing, and green cloud environment testing to assure cloud-based program quality. The first year of this project is to design the models, including deployment model, service composition model, as well a testing model, so that the analysis to assure the requested requirements on QoS and SLAs can be applied to these models. The design and development of methodologies and prototyping tools will be implemented. The main efforts of the second year will be on the integration of the sub-projects and experiments on some industry projects, such as migrating ERP, CRM systems to cloud computing. The third year of this project will apply our integrated solutions to real applications from of the software industry to demonstrate the feasibility of our project. The implementation of this project will be based on open source. |