Efficient partitioning of graphs in collaborative workflow editor systems

Collaborative editing systems allow a group of users to view and edit a shared item from geographically dispersed sites. Consistency maintenance in the face of concurrent accesses to shared entities is one of the core issues in the design of these systems. Workflow modelling is a popular technique to describe business processes, scientific experiments, distributed applications. A workflow is directed graph which specifies tasks and data/control dependencies. The paper introduces protocols by which workflow developer environments can enable the concurrent editing of graphs by multiple users. The proposed graph partitioning and pessimistic locking algorithms assure that collaborators cannot break the consistency criteria of workflows by introducing cycles or invalid edges to them. We prove that the solution results correct graphs even when collaborative parties know separate parts of the workflow and do not share their own sub-graphs with each other in real time. A method to compare the efficiency of different graph partitioning algorithms is also provided