Towards a Provisioning Algorithm for Dynamic Workflows in the Cloud

Workflow applications are an enabling technology for coordinating the activities of an enterprise. The resources provisioning for these applications is one of the most difficult challenges in the cloud. It consists in making an appropriate decision when mapping tasks to resources while satisfying QoS requirements. Current resources provisioning approaches in the cloud consider only static workflows and ignore the need to change workflow instances at runtime. This functionality is an essential requirement to deal with unusual situations and evolutions. In this paper, we propose a novel resources provisioning approach which considers the dynamic changes of workflows. The proposed approach introduces an algorithm which determines the assignment of tasks to the appropriate cloud resources. After that, we extend it in order to support the dynamic changes of workflows. Our algorithm is evaluated using CloudSim simulator. The experimental results which we present illustrate the efficiency of our approach in term of financial execution cost.