Transparent data exchange in service choreographies : an eScience perspective

Abstract

This work is motivated by the increasing importance and business value of data in the fields of business process management, scientific workflows as a field in eScience, and Internet of Things, all of which profiting fromthe recent advances in data science and Big data. Although service choreographies provide means to specify complex conversations betweenmultiple interacting parties from a global perspective and in a technology-agnostic manner, they do not fully reflect the current paradigm shift towards data-awareness at the moment. Therefore, the focus of this work is on tackling respective shortcomings. These include the missing modeling support for data flow across participant boundaries or specifying a choreography data model as a contract on the business data relevant to realize the collaboration and all interacting parties agree on. Towards this goal, we introduce a choreography management life cycle that assigns data its deserved primary role in service choreographies as well as defines the functions and artifacts necessary for enabling transparent and efficient data exchange among choreography participants. To implement the introduced life cycle we present the notion of data-aware choreographies through our concepts for Transparent Data Exchange (TraDE) by introducing cross-partner data objects and cross-partner data flows as means to increase runtime flexibility while reducing the complexity of modeling data flows in service choreographies. The TraDE concepts focus on decoupling the data flow, data exchange and management, from the control flow in service compositions and choreographies. To provide an end-to-end support for the modeling and execution of data-aware choreographies and supporting the respective phases of the choreography management life cycle, we introduce and prototypically implement an overall TraDE ecosystem. This ecosystem comprises a modeling environment for data-aware choreographies as well as the required runtime environment to execute such data-aware choreographies through a new TraDE Middleware and its integration to corresponding Business Process Engines (BPEs). The inherent goal of this work is to simplify the modeling of data and its exchange in choreography models while increasing their runtime flexibility and enabling the transparent exchange and transformation of data during choreography execution.