Enhancing workflow fault tolerance using replication technique

by 1974- Zhu, Xiaoliang

ORBWork is a scalable distributed runtime system of the METEOR workflow management system. This thesis investigates providing fault tolerance through replication in workflow systems. The passive and active replication models are implemented using Sun Jini technology and applied to ORBWork. Our experiments show that replication can greatly enhance the fault tolerance of the workflow system. Especially, the system can tolerate fatal failures including server crashes. Index words: Workflow, Replica, Fault Tolerance, Computer, Reliability, Availability, Exception, Java, RMI, Jini Enhancing workflow fault tolerance Using Replication Technique by Xiaoliang Zhu B.S., The University of Science and Technology of China, China, 1998 A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree Master of Science Athens, Georgia 2002 c? 2002 Xiaoliang Zhu All Rights Reserved Enhancing workflow fault tolerance Using Replication Technique by Xiaoliang Zhu Approved: Major Professor: Krzysztof J. Kochut Committee: Amit P. Sheth Daniel M. Everett Electronic Version Approved: Gordhan L. Patel Dean of the Graduate School The University of Georgia August 2002