PROJECT TITLE :

Runtime Verification of Business Cloud Workflow Temporal Conformance

ABSTRACT:

Many times, business workflows in the cloud are designed with multiple time constraints in order to respond to business requests in a timely manner. In order to guarantee that workflow instances are finished on time, the workflow's temporal conformance state must be continuously monitored and verified while the workflow is being executed. Conventional verification approaches for time-related properties based on temporal logic or timed Petri nets are not feasible because of the limitation of low efficiency at runtime. This is due to the fact that many business scenarios involve a large number of workflow instances that are running in parallel with one another. In order to solve this problem, we have come up with a novel strategy for the automated runtime verification of temporal conformance for parallel workflow instances operating in a cloud-based environment. As opposed to using response time as the performance measurement to verify the temporal conformance of each and every workflow, as is done in traditional monitoring strategies, workflow throughput is used as the performance measurement in this article to effectively monitor a large number of parallel workflow instances. On the basis of this, we present an innovative strategy for conformance verification. In order to accurately verify the workflow's runtime temporal conformance, this strategy takes into account the effect of time delay propagation within the cloud workflow systems. The results of the evaluation show that our verification strategy is superior to the workflow temporal verification strategy that is currently considered to be state-of-the-art. This strategy has been implemented in a prototype cloud workflow system.