Scheduling Data-Driven Workflows in Multi-cloud Environment

Nowadays, cloud computing and other distributed computing systems have been developed to support various types of workflows in applications. Due to the restrictions in the use of one cloud provider, the concept of multiple clouds has been proposed. In multiple clouds, scheduling workflows with large amounts of data is a well-known NP-Hard problem. The existing scheduling algorithms have not paid attention to the data dependency issues and their importance in scheduling criteria such as time and cost. In this paper, we propose a communication based algorithm for workflows with huge volumes of data in a multi-cloud environment. The proposed algorithm changes the definition of the Partial Critical Paths (PCP) to minimize the cost of workflow execution while meeting a user defined deadline.