SaaS系统管理与进化自主算法研究
[Abstract]:SaaS (Software as a Service) is the provision of software through the Internet. Unlike traditional software models, software that uses the SaaS model does not need to buy software products and install them on personal computers. SaaS software is growing exponentially, and more enterprises of all sizes are using SaaS model software. In order to improve performance and save costs. This type of application is therefore a more attractive economic model than the software permanent licensing model. The goal of autonomous computing is to develop an application software system which can realize self-management based on the high-level guidance of human beings. There are many challenges in the research on the evolution of SaaS application software and the improvement of quality of service (QoS) autonomous algorithms. In order to realize the self-management and evolution of SaaS application software, a meta-model of SaaS application software is proposed in this paper. This metamodel describes the multi-tier system structure of SaaS application software, which is from bottom to top: database layer, service layer, business process layer and user interface. From the point of view of SaaS application software provider, tenant and user, this paper describes the relationship and dependence among the components of each layer. In order to enable SaaS applications to dynamically adapt to the changing needs of SaaS providers, tenants, and users, the service applicability requirements and QoS management requirements of different levels of SaaS applications are addressed. The formal definition and description are carried out by using PCTL (Probabilistic Computation Tree Logic) and CSL (Continuous Stochastic Logic) description language, and two kinds of quality evaluation models of discrete SaaS user-tenant-provider model (DSUTP) and continuous SaaS user-tenant-provider model (CSUTP) are proposed. It is used to analyze and evaluate SaaS application software. Based on the metamodel and model of SaaS application software, an autonomous algorithm AAQS, for QoS management of SaaS application software is proposed to improve the quality of service of SaaS application software. The algorithm can monitor and predict the QoS violation of SaaS application software. The AAQS algorithm is evaluated by practice and system analysis. In order to enhance the evolutionary ability of SaaS application software, an autonomous evolutionary algorithm (AAES).) is proposed for SaaS application software. Combined with meta-heuristic method and case-based reasoning (CBR) method, the SaaS program selection method based on SaaS case-based reasoning (CBR) is studied. SaaSCBR.SaaSCBR is an appropriate SaaS application software based on best practice case selection. The experimental results show that the AAES algorithm improves the performance of SaaS application software by selecting SaaS application cases suitable for SaaS tenants, tenants and users. Finally, combined with the SaaS QoS management and evolution method, the autonomous management algorithm (AASS), of SaaS system is proposed to realize the self-management and evolution of SaaS system. A practical case study of autonomous SaaS system based on EHR system SaaSEHR, based on SaaS is developed. SaaSEHR uses prior and posterior knowledge to realize autonomous evolution. The experimental results show that by selecting SaaS services suitable for SaaS tenants and users, AASS can improve the self-optimization and self-healing ability of the SaaS system, and enhance the adaptability of the SaaS system to changes in environment and requirements. Moreover, the performance of SaaS system and the ability to control SaaS services are improved by sharing business processes between SaaS tenants and users.
【学位授予单位】:哈尔滨工业大学
【学位级别】:博士
【学位授予年份】:2014
【分类号】:TP393.09
【共引文献】
相关期刊论文 前10条
1 朱彭辉;张振明;黄利江;;基于实例推理的车间生产调度方法研究[J];制造业自动化;2014年05期
2 黄树成;罗志;陈建;;面向开放环境的自适应软件构造方法[J];江苏科技大学学报(自然科学版);2014年01期
3 陈江涛;张金隆;张亚军;;基于云计算的IT服务定价研究综述及前景展望[J];科技管理研究;2014年10期
4 杨荣;李兵;杜宝同;熊伟;何鹏;;满足原子事务的QoS感知的自适应服务选择[J];计算机科学与探索;2014年05期
5 周笑波;蒋昌俊;;Autonomic Performance and Power Control on Virtualized Servers:Survey, Practices, and Trends[J];Journal of Computer Science & Technology;2014年04期
6 曾显华;陈建州;刘立祥;;GEO通信系统中MAC层信道分配算法研究[J];计算机应用与软件;2014年06期
7 李晓明;姜晓曦;;云计算对图书馆基本业务框架的影响研究[J];图书馆建设;2013年10期
8 林俊宇;王慧强;马春光;卢旭;吕宏武;;一种基于DAG动态重构的认知网络服务迁移方法[J];软件学报;2014年10期
9 卞艺杰;马玲玲;;云环境下移动视频监控系统安全性研究[J];计算机技术 与发展;2013年09期
10 柯平;朱明;闫娜;;国外图书馆管理研究述评[J];中国图书馆学报;2013年05期
相关博士学位论文 前6条
1 金彦龄;基于生物智能的物联网协同自治机理研究[D];东华大学;2013年
2 杨冬;Web服务中若干问题的研究[D];吉林大学;2014年
3 张鹏;形式化方法在云计算中的应用研究[D];吉林大学;2014年
4 杨英;LTE网络自优化方法研究[D];中国科学技术大学;2014年
5 王士显;面向认知无线电的数字信号处理器体系结构技术研究[D];国防科学技术大学;2013年
6 孟超;基于云计算的病毒恶意软件分析研究[D];南京航空航天大学;2013年
相关硕士学位论文 前10条
1 贾英霞;高校图书馆新技术应用发展对策研究[D];河北大学;2013年
2 李化;智能移动服务开发方法的应用研究[D];北京工业大学;2013年
3 鞠立涛;面向部队非密物资采购服务的服务平台设计与实现[D];哈尔滨工业大学;2012年
4 周柳英;湖南省图书馆联盟现状与发展对策研究[D];中南大学;2013年
5 刘舒;Android功能自动化测试工具的设计与实现[D];大连理工大学;2013年
6 刘松涛;小区PCI冲突和混淆的研究[D];武汉邮电科学研究院;2012年
7 谢南;支持ASBS自适应的服务虚拟社区动态调整方法[D];东北大学;2013年
8 左卓;基于ILASII系统的图书管理系统研究[D];湖南大学;2013年
9 曹帅;面向产业链协同SaaS平台的流程定制系统研究[D];西南交通大学;2014年
10 侯德智;多租户数据放置与访问研究[D];山东大学;2014年
本文编号:2471408
本文链接:https://www.wllwen.com/guanlilunwen/ydhl/2471408.html