ABSTRACT
The concept of workflow management system originates from the field of office
automation. The workflow passes the task or information between the participants
according to some defined rules. The workflow refers to the combination of the ideas,
theories, methods, skills and systems that support business processes to run efficiently,
and also is the key technology of the enterprise business process reengineering and
business process automation. In the current complex, heterogeneous, distributed
enterprise environments, distributed workflow management system is becoming a hot
direction of research. However, people are more concerned about the process modeling
of workflow and scheduling of task etc。The data stream research of the workflow is
fragmented. This paper considers data management in the workflow system to be a main
research subject, focusing on three areas of data management in distributed workflow
environment: heterogeneous data interaction, data consistency protection, adaptive
allocation of resource data.
Firstly, the paper introduces the research background and current status of the
research. Secondly, a number of key technologies about our research are introduced.
Finally, we focus on three aspects of the workflow management's data management. The
specific research content is as follows:
1. The interaction and integration of heterogeneous data in distributed workflow
environment: Firstly, we introduce various heterogeneous data in distributed workflow
environment and classify the data according to the characteristics. Then the middleware
ideas and the XML technology were introduced in workflow environment. The main
contribution of this paper is to construct middleware-based interaction architecture and
define the overall pattern of the control data and associated data using xml schema.
2. Consistency protection of the workflow data: Firstly, this paper analyzes two
important reasons which destroy the data consistency in workflow system: exception
and concurrency. Then, for exceptions we propose the event-based multi-version control
algorithm to recover the data; and this paper also presents transactional workflow mixed
concurrency control based on semantic isolation sphere to solve the data problems
arising from concurrency.
3. The adaptive distribution of resource data: Various resource data dynamically