From Centralized Workflow Specification to Distributed WorkflowExecution

  • Authors:
  • Peter Muth;Dirk Wodtke;Jeanine Weissenfels;Angelika Kotz Dittrich;Gerhard Weikum

  • Affiliations:
  • Department of Computer Science, University of the Saarland, P.O. Box 151150, D-66041 Saarbrü/cken, Germany/ www: http://www-dbs.cs.uni-sb.de/&sim/mentor.;Department of Computer Science, University of the Saarland, P.O. Box 151150, D-66041 Saarbrü/cken, Germany/ www: http://www-dbs.cs.uni-sb.de/&sim/mentor.;Department of Computer Science, University of the Saarland, P.O. Box 151150, D-66041 Saarbrü/cken, Germany/ www: http://www-dbs.cs.uni-sb.de/&sim/mentor.;Union Bank of Switzerland, Bahnhofstrasse 45, CH-8021 Zü/rich, Switzerland. E-mail: kotz-dittrich@ubs.ch;Department of Computer Science, University of the Saarland, P.O. Box 151150, D-66041 Saarbrü/cken, Germany. E-mail: weikum@cs.uni-sb.de

  • Venue:
  • Journal of Intelligent Information Systems - Special issue on workflow management systems
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

Current workflow management systems fall short ofsupporting large-scale distributed, enterprise-wide applications. Wepresent a scalable, rigorously founded approach to enterprise-wideworkflow management, based on the distributed execution of state andactivity charts. By exploiting the formal semantics of state andactivity charts, we develop an algorithm for transforming acentralized state and activity chart into a provably equivalentpartitioned one, suitable for distributed execution. Asynchronization scheme is developed that guarantees an executionequivalent to a non-distributed one. This basic solution is furtherrefined in order to reduce communication overhead and exploitparallelism between partitions whenever possible. The developedsynchronization schemes are compared in terms of the number and sizeof synchronization messages.