Stochastic Model Checking Framework for Complex Cloud Applications

来源 :2015全国理论计算机科学学术年会 | 被引量 : 0次 | 上传用户:henrychan168
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  In virtualized and dynamical cloud computing environment, all resources such as infrastructure, hardware,platform, software and data can be virtualized and partitioned into some kinds of resouces pools and provided as IT services which can be accessed through internet in a pervasive way.One can create new value-added cloud applications by aggegating these virtualized resouces at different levels of abstraction.Response time and service cost of an aggregated cloud application are the key issue which users concern about mainly.Therefore, in design phase, we are supposed to verify whether the functionality together with the constraints of response time and service cost of cloud applications fulfill users requirements.Unfortunately, some existing approaches treat them separately.This paper presents the modeling framework for cloud applications and takes asMRM which is extended from Markov Reward Models as the formal model.asMRM has action and state labels and can depict stochastic behavior that cloud applications exhibit due to dynamicity and nondeterminacy.We introduce the logic asCSRL which provides a powerful means to characterize specifications which are equipped wvith time-interval-as well as cost-intervalbounds.The model checking procedue for asCSRL boils down to model checking CSRL formula in Markov reward model with the help of model checker MRMC.
其他文献
  Modern search engines process thousands of queries per second, searching through billions of web documents.With data sizes and query loads growing at an exp
会议
  通过在移动医疗大数据平台下,机会网络中节点传递信息方式的特点分析,遍历所有邻接节点,对两节点的数据进行比较,通过最优匹配方式,选择匹配结果最优的邻接节点作为下一
会议
  分类是数据挖掘和数据分析中最有应用价值的技术之一.传统的积极学习方法需要预先对模型空间进行假设,并且没有充分考虑到实例之间的相关性,其泛化能力将会受到一定程度的
会议
  In the progress of fluid simulation by particle method, the particles is disordered, so that it is difficult to achieve surface extracting by tracking the p
会议
  机会网络的数据交换不需要完整的路径,且传递数据是基于节点机会移动和机会相遇。相对于传统网络来说,达到了绿色节能的目的。但是机会网络中节点逐跳传递数据存在较大的传
会议
  轻度认知功能障碍是介于大脑正常老化和老年痴呆的一种认知缺损状态,随着年龄增长,MCI发生率随之增长趋势明显。中医认为本病病机为本虚标实,虚实夹杂。灸法作为中国传统
异常点检测在机器学习和数据挖掘领域中有着十分重要的作用.当前异常点检测算法的一大缺陷是正常数据在边缘处异常度较高,导致在某些情况下误判异常点.为了解决该问题,提出了
  单纯性肥胖症是一种临床较为常见的疾病,它严重影响患者的外观形象及心理,病情反复,治愈困难。已有研究表明,针灸治疗单纯性肥胖症具有其独特的优势,本文通过对近年来针灸治疗
  化疗有较多的副反应,严重影响患者生存质量及化疗疗效。针灸治疗具有双向良性调节作用,减少化疗后副反应,临床研究疗效确切,且不良反应少,应用方便,为临床工作者提供有效治疗方
  支气管哮喘的致病和诱发因素比较复杂,心理因素也是其中之一。哮喘作为一种心身疾病已经成为医学界的共识。然而在临床上,面对病人时,医生通常更加重视患者的症状而忽视其心