Title :
A Continuous Long Running Batch Orchestration Model for Workflow Instance Migration
Author :
Frank, David ; Fong, Liana ; Lam, Linh
Abstract :
Workflow engines which require process instances to run to completion against their creation model are limited in their ability to adapt to model changes. The technique of process instance migration can be used to migrate instances to a new process model, but at any given time only a fraction of instances may be in a state which is compatible with the new model. We propose a continuous batch approach for migration which treats migration itself as a long running problem. The continuous migration approach schedules and orchestrates the evaluation and migration of long running processes over time, controlled by configurable policies while maintaining continuous process availability. We have built a long running process migration simulation to test our approach, looking at both migration success and cost. The simulation confirms the need for a long running migration approach and yields some interesting suggestions for tuning the configuration of the migration system.
Keywords :
business data processing; workflow management software; batch approach; batch orchestration model; long running process migration simulation; workflow engines; workflow instance migration; Adaptation model; Analytical models; Availability; Business; Flowcharts; Heuristic algorithms; Process control; Migration; business processes; long-running; process instances; workflow; workflow engines;
Conference_Titel :
Services Computing (SCC), 2010 IEEE International Conference on
Conference_Location :
Miami, FL
Print_ISBN :
978-1-4244-8147-7
Electronic_ISBN :
978-0-7695-4126-6
DOI :
10.1109/SCC.2010.90