分布式环境下业务模型的数据存储及访问框架

发布时间：2018-04-18 02:41

本文选题：云服务系统 + 业务模型　；参考：《清华大学学报(自然科学版)》2017年06期

【摘要】：构造业务模型以支持应用系统开发是基于模型驱动架构实现云服务系统快速开发及有效运维的重要途径。然而,云平台下的海量异构模型的统一管理并不容易。该文提出一个分布式环境下的业务模型数据存储访问框架,以支持业务建模、服务转换、服务配置、服务部署、服务监控等服务生命周期。将关系数据库及NoSQL数据库结合以存储和管理结构化数据;采用基于Hadoop构建的文件库以实现非结构化的业务模型管理,从而构造一个综合的数据管理模型,实现了业务模型统一存储和管理;并根据业务模型中的资源描述,生成相应RESTful服务供应用系统调用;构建一个基于云平台的业务模型库软件平台以开展应用验证。结果表明:该框架不仅对云服务应用系统提供了高效的数据存储访问方式,也降低了应用服务的开发及维护成本,具有较高工程价值。
[Abstract]:Constructing business model to support application system development is an important way to realize rapid development and effective operation and maintenance of cloud service system based on model-driven architecture.However, the unified management of massive heterogeneous models on cloud platform is not easy.In this paper, a data storage access framework for business model in distributed environment is proposed to support service lifecycle such as business modeling, service transformation, service configuration, service deployment, service monitoring and so on.The relational database and the NoSQL database are combined to store and manage the structured data, and the document library based on Hadoop is used to realize the management of the unstructured business model, thus a comprehensive data management model is constructed.It realizes the unified storage and management of the business model, generates the corresponding RESTful service provisioning system calls according to the resource description in the business model, and constructs a business model base software platform based on the cloud platform to carry out application verification.The results show that this framework not only provides an efficient data storage and access mode for cloud service application systems, but also reduces the development and maintenance costs of application services, and has high engineering value.
【作者单位】：上海交通大学软件学院;
【基金】：国家自然科学基金面上项目(61373030,71171132)
【分类号】：TP393.09

【相似文献】