On Evolving Partitioned Web Service Orchestrations

Many researches argue that centralized Web Service (WS) orchestrations stop short in dealing with key requirements such as scalability, privacy and reliability. Consequently, fragmentation and decentralization have been proposed to overcome these limitations. In detail, the centralized orchestration is fragmented into behaviorally equivalent distributed partitions such that their combined execution recreates the function of the original orchestration. However, the evolving nature of business processes created the need for an efficient change support. Since the decentralization leads to the distribution of the activities, the control and data flows, it becomes difficult to specify the changes directly on the derived partitions. Therefore, it is more judicious to specify the changes on the centralized orchestration model and propagate them to the derived partitions. In this paper, we propose a comprehensive change framework for partitioned WS orchestration scenarios and demonstrate how to specify and propagate the changes from the centralized model to its resulting decentralized partitions.