Job Allocation Using Two-Level Prioritization Stage Directed Acyclic Graph (DAG) Model in Cloud Environment

Cloud computing has emerged as a popular computing paradigm for hosting large computing systems and services. Several significant studies have been carried out on job allocation; these studies focus on efficient sharing of cloud resources among multiple users. Considering the high volume of users' demands and the diversity of resource capacities, efficient job allocation has become necessary and challenging especially in scientific and complex applications. In this paper, we propose a dynamic job allocation approach for the directed acyclic graph application model by using two-level prioritization stages in a cloud environment. Our allocation approach explicitly considered priority list and identified job ranking when making allocation decisions. Simulation results demonstrate that our proposed approach is able to improve makespan and schedule length.