Modeling Platform Design Based On Process Engine

SOA because of its loose coupling, cross-platform have recently been widely used. Traditional SOA architecture generally is two-tier, from demand to components, middle semantic layer defaults to XML, but this is only related to the field of software. The laboratory proposed three-tier SOA architecture from traditional SOA architecture, and the middle semantic layer mapped into process, basing on LL7semantic description proposed by laboratory.Traditional banking business platform based on system requirements, lacking of a detailed business description. In this thesis the bank card business modeling as example, divided its three-tier SOA business architecture model, conceptual model corresponds to the type of business areas, business areas, the type of business use cases, business use cases, business use case scenario, logical model corresponds to a flow chart, including a variety of business steps, physical model corresponds to the member that is encapsulated between the atomic steps and between the atomic steps and basic steps, where the atomic step has independence and completeness, that the atomic step is not in the segmentation step, and step may be combined to generate by the different atomic steps.Through the bank card business modeling and then extended to the general business modeling, modeling to modeling platform. General business modeling platform modeling correspond to three-layer model of the SOA architecture, the conceptual model demand corresponding engine, the semantics of the logical model corresponds process, including the sequence of steps, criteria steps, cycle steps; physical model of service corresponds member, including process body, basic steps and atomic steps or atomic steps, general business modeling platform for all business conducted three SOA architecture modeling, extract its atomic components.Process engine, by analyzing51IP check to understand LL7control section and operating atomic step package sql statement, and complete the call to the component.Finally, summarize, and further research direction and content.