Research and Implementation of Distributed Workflow System Based on Shark System
|Course||Applied Computer Technology|
|Keywords||enhydra shark Distributed Workflow Flexible execution Data consistency|
Relatively poor prevalent in the workflow management system flexibility , robustness and interoperability problems , the paper combines the experience of actual development workflow , a flexible , interoperability, multi- engine distributed workflow systems. The system is based on the single-engine Shark workflow system is modeled to build a distributed workflow management system (DWFMS) architecture , called D-shark (distributed shark) workflow management system . D-shark retain the original workflow engine for local processing , a master engine to manage the distribution of the model , the console services as well as remote access . System underlying communication mechanism using standard message communication , to ensure the reliability of system interoperability between . The increase in D-shark system node distributor , activity model in order to be assigned to the other engine up execution to achieve interoperability . The paper presents the concept of separation of activities and control logic , given the dynamic modification of workflow model strategy , dynamic modeling of the D-shark . Papers exception handling , data consistency problems exist in the system of D-shark -depth study , and given the appropriate solution .