Abstract: For the purpose of solving the problem of Information Island in distributed heterogeneous data sharing, the authors designed and implemented DataTurbo, a platform used to migrate, integrate and syncretize distributed data. DataTurbo was considered to be a comprehensive platform: by the advantage of scalable component design, it supports userlevel strategy description; with flexible scheduling engine, it covers commercial softwares weakness that it only supports normalized data access interface and has limited functions. The article emphasized the analysis of framework of DataTurbo, the design and implementation of the sustaining level below the function level, and description about the core of scheduling logic and the specification of function component. It has been installed and served in the data center for Panyu District of Guangzhou successfully. DataTurbo can afford a great amount of heterogeneous data exchange and synchronization and is suitable to centralize data between departments in governments and corporations.

Key words: heterogeneous data

摘要： 针对分布式异构数据共享存在的“信息孤岛”问题，设计并实现了一个用于解决分布式数据迁移、集成、融合的平台DataTurbo。DataTurbo利用可扩展功能构件支持贴近用户语义的策略描述，结合灵活的调度引擎，弥补了很多商业工具仅支持规范化的数据访问接口以及功能局限的弱点，成为一个综合的平台。重点剖析了DataTurbo系统的框架结构，强调功能层下面的引擎支撑层的设计与实现，重点描述了核心调度逻辑和功能构件的规范。DataTurbo现已成功部署并服务于广州市番禺区数据中心，能承担大量异构数据的交换与同步，适合政府各级部门、企业各级机构的数据集成、共享与数据中心的构建。

关键词: 数据集成, 可扩展, 引擎, 异构数据, 策略, data integration, extensible, engine, strategy