A decentralized execution model for inter-organizational workflows

Abstract

Workflow Management Systems (WFMS) are often used to support the automated execution of business processes. In today's networked environment, it is not uncommon for organizations representing different business partners to collaborate for providing value-added services and products. As such, workflows representing the business processes in this loosely-coupled, dynamic and ad hoc coalition environment tend to span across the organizational boundaries. As a result, it is not viable to employ a single centralized WFMS to control the execution of the inter-organizational workflow due to limited scalability, availability and performance. To this end, in this paper, we present a decentralized workflow model, where inter-task dependencies are enforced without requiring to have a centralized WFMS. In our model, a workflow is divided into partitions called self-describing workflows, and handled by a light weight workflow management component, called the workflow stub, located at each organization. We present a performance study by considering different types of workflows with varying degrees of parallelism. Our performance results indicate that decentral-56 Distrib Parallel Databases (2007) 22: ized workflow management indeed enjoys significant gain in performance over its centralized counterpart in cases where there is less parallelism.

Synchronising Service Compositions in Dynamic Ad Hoc Environments

“…We adopt a methodology similar to the one Atluri et al [25] used and test seven requests types with seven services that vary in the number and length of AND-splitting We compare early and late runtime binding and embed each in a composition model. CiAN, a workflow engine for mobile ad hoc networks [4], implements early binding in addition to beacon-based service advertisement, centralised fragmentation-based provider allocation, and decentralised service execution.…”

A Decentralized Approach to Network-Aware Service Composition

“…The CO model simplifies the workflow management, but introduces additional delays caused by the indirect interaction between consecutive services, and may suffer from the typical problems of a centralized solution (bottleneck node, single point of failure). The DO model overcomes these problems, but requires the instantiation of workflow control logic at each node hosting a workflow resource [2,4]. This can be problematic, as such nodes could not be willing to host this logic, or could have limited capabilities that make them not capable of coordinating the workflow operations (as could be the case of nodes involved in Internet-of-Things scenarios).…”

Migratability of BPMN 2.0 Process Instances

“…The control of a particular process instance migrates from one execution server [14] to another and process activities can influence the next participant. Related to this, Atluri et al [1] present a process partitioning algorithm which creates self-describing subprocesses allowing dynamic routing and assignment. Process migration has also particularly been applied to the area of mobile process execution, e.g.…”

“…[8,13]) or use a completely proprietary specification (e.g. [6,10,4,1]). Based on its novelty as an executable process description language, migration of (unmodified) BPMN process instances has not been considered yet.…”