Research And Implementation Of Distributed Data System Based On Service Creation Platform

Service creation platform provides a set of tools to encapsulate business entities into atomic services using SOA technologies. Based on service creation platform, system users could make quick combination of separate services and reuse resources for system integration by encapsulating distributed functional units into standard services in a uniformed way. There are lots of difficulties in managing mixed services in heterogeneous environment currently. Service creation platform is a solution to heterogeneous problems of enterprise.With rapid development of computer technology and service applications, distributed technology becomes widely popular in setting up enterprise-class applications. Traditionally, centralized data storage is too simplistic, and a single server accumulates large amounts of data, resulting in a decline of response rate or even collapse of the server. In addition, the failure of a single server will lead to the loss of large amounts of data. That kind of situation causes business losses, or even more serious business failures. What’s more, Centralized data storage can’t solve the problem of off-site storage. Some large-scale companies have a lot of scattered subsidiaries around. However, traditional data storage will bring about much waste of manpower and funds.This paper investigates and implements distributed data service system which serves service creation platform. The system provides a new distributed architecture, storing data to different data nodes by distributed agent; The system provides a new distributed management framework, using metadata server to manage other function modules; The system provides a new data segmentation schema, storing business data to different data nodes according to user-defined configuration.Firstly, related technical background is introduced. Secondly, a detailed description of the architecture of distributed data service system is introduced. What’s more, four functional modules of the distributed data system are separately presented from the perspective of design and implementation. Finally, further improvement of distributed data service system in future is recommended.