【摘要】：在各類云服務(wù)呈現(xiàn)爆炸式增長的今天，用戶對(duì)云服務(wù)的穩(wěn)定性與健壯性的要求也越來越高，在將云服務(wù)提供給用戶使用之前，云服務(wù)提供商需要對(duì)服務(wù)進(jìn)行全面的測試以保證服務(wù)的質(zhì)量，而在各類云服務(wù)的測試中，其健壯性測試是不可缺少的。 本文研究了當(dāng)前主流健壯性測試方法，根據(jù)其測試對(duì)象的不同對(duì)其測試方法進(jìn)行了研究與分析對(duì)比，在綜合了不同層次的云服務(wù)特點(diǎn)后，提出了一套針對(duì)不同層次云服務(wù)的健壯性測試與度量的方法。 在針對(duì)具體的IaaS、SaaS、PaaS三層云服務(wù)的研究基礎(chǔ)上，，我們選擇支撐各層服務(wù)的核心技術(shù)或有代表性的云服務(wù)對(duì)其進(jìn)行健壯性測試，從測試方便性與用戶使用范圍的角度考慮，我們優(yōu)先選擇開源的云服務(wù)核心技術(shù)作為測試對(duì)象。其中，IaaS層我們選擇利用開源的CloudStack云計(jì)算解決方案而搭建的IaaS云服務(wù)進(jìn)行測試，PaaS層選擇為多種PaaS云服務(wù)提供重要技術(shù)支持的Hadoop開源分布式計(jì)算架構(gòu)進(jìn)行測試，SaaS層選擇典型的SaaS層云服務(wù)SalesforceCRM進(jìn)行測試。 為了測試的方便與高效快捷，我們對(duì)各層云服務(wù)的測試進(jìn)行了集成，搭建了云服務(wù)健壯性測試平臺(tái)，平臺(tái)實(shí)現(xiàn)了與測試人員的界面交互與測試對(duì)象的選擇功能，并負(fù)責(zé)收集測試數(shù)據(jù)進(jìn)行統(tǒng)計(jì)。該測試平臺(tái)可以很好的兼容各類云服務(wù)的健壯性測試工具，在實(shí)現(xiàn)代表云服務(wù)的測試集成后，后續(xù)服務(wù)的健壯性測試可以很方便的添加到該測試平臺(tái)下，兼容性高。 根據(jù)不同云服務(wù)的評(píng)測指標(biāo)，我們對(duì)測試對(duì)象提供的接口進(jìn)行了分類，分別對(duì)其測試結(jié)果進(jìn)行統(tǒng)計(jì)并計(jì)算其失效率，根據(jù)不同類別接口導(dǎo)致的服務(wù)失效率的比對(duì)結(jié)果對(duì)其云服務(wù)的健壯性進(jìn)行評(píng)估，指出云服務(wù)設(shè)計(jì)上可能存在的導(dǎo)致不健壯性因素的薄弱環(huán)節(jié)。在完成各類接口測試的基礎(chǔ)上，按照接口權(quán)值對(duì)云服務(wù)進(jìn)行失效率評(píng)估，對(duì)不同層次的云服務(wù)健壯性進(jìn)行對(duì)比分析與總結(jié)。
[Abstract]:With the explosive growth of all kinds of cloud services, users are demanding more and more stability and robustness of cloud services. Cloud service providers need to conduct a comprehensive test to ensure the quality of services, but in the testing of all kinds of cloud services, its robustness test is indispensable. In this paper, the current mainstream robust testing methods are studied, and the test methods are compared and analyzed according to the different test objects. After synthesizing the characteristics of different levels of cloud services, This paper presents a method of robustness testing and measurement for cloud services at different levels. On the basis of the research on the specific IaaS,SaaS,PaaS three-layer cloud service, we choose the core technology to support each layer of service or the representative cloud service to carry on the robust test to it, consider from the point of view of the test convenience and the scope of the user's use. We give priority to open source cloud service core technology as test object. Among them, IaaS layer we choose to use open source CloudStack cloud computing solutions to build IaaS cloud services for testing, PaaS layer choose to provide important technical support for a variety of PaaS cloud services for the Hadoop open source distributed computing architecture to test, SaaS layer selection Test the typical SaaS layer cloud service SalesforceCRM. In order to test conveniently and efficiently, we integrate the test of each layer cloud service, and build the test platform of cloud service robustness. The platform realizes the interface interaction with testers and the function of selecting test objects. And responsible for collecting test data for statistics. The test platform can be well compatible with all kinds of cloud services robustness testing tools. After implementing the test integration on behalf of cloud services, the robust test of subsequent services can be easily added to the test platform, which has high compatibility. According to the evaluation indexes of different cloud services, we classify the interfaces provided by the test objects, and calculate the failure rate of the test results. According to the comparison results of service failure rate caused by different types of interfaces, the robustness of cloud services is evaluated, and the weak links that may lead to unrobustness factors in the design of cloud services are pointed out. On the basis of all kinds of interface tests, the failure rate of cloud services is evaluated according to the interface weights, and the robustness of cloud services at different levels is analyzed and summarized.
【學(xué)位授予單位】：哈爾濱工業(yè)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP393.09

【參考文獻(xiàn)】

相關(guān)期刊論文 前6條

1 白秀杰;李汝鑫;劉新春;邵宗有;;云安全防護(hù)體系架構(gòu)研究[J];信息安全與技術(shù);2013年05期

2 ;Amazon云計(jì)算平臺(tái)Web Services和MySQL的碰撞[J];硅谷;2011年21期

3 姜麗娟;錢步仁;;基于XML的協(xié)議健壯性測試工具研究與設(shè)計(jì)[J];科學(xué)技術(shù)與工程;2010年14期

4 李喬;柯棟梁;王小林;;云測試研究現(xiàn)狀綜述[J];計(jì)算機(jī)應(yīng)用研究;2012年12期

5 張鵬;;國內(nèi)云產(chǎn)業(yè)縱深發(fā)展 三大優(yōu)勢引運(yùn)營商IDC青睞[J];通信世界;2013年24期

6 肖軍;胡雷鳴;;基于錯(cuò)誤的Web服務(wù)魯棒性測試方法研究[J];微計(jì)算機(jī)信息;2010年31期

本文編號(hào)：2287305