User priority aware and cost constrained workflow scheduling in clouds

Cloud computing can provide powerful and reliable services for the increasing complex applications. The application workflows can be migrated to the cloud resources for processing. Most existing meta-heuristic and heuristic algorithms concentrate on orchestrating workflows under users' Quality of Service parameters, nevertheless, the user priority is usually neglected. In this paper, we design a new scheduling algorithm named User Priority Aware and Cost Constrained Workflow Scheduling (UPCWS) to execute multiple applications arriving at the same instant in heterogeneous distributed systems, which takes both user priority and QoS parameters into consideration. The algorithm decomposes the global workflow demands into sub-constraints of tasks and filters all available processors to obtain the suitable resources which satisfy the cost constraints. The algorithm of UPCWS can automatically determines the execution order of tasks by pre-scheduling and real time adjustment according to the priority of workflows. Experimental results indicate that the UPCWS algorithm can obtain higher planning success rate comparing against several state-of-the-art algorithms with a smaller monetary cost. It's also proved that the algorithm yields the best performance for the high priority users in the situation of changing the number of workflows.