Delft University of Technology
A growing number of applications are running in the cloud. Academia and industry are both increasingly using cloud resources as infrastructure to serve their users, due to the elastic, flexible, and pay-as-you-go features of Infrastructure-as-a-Service (IaaS) clouds. Cloud brokers need to lease resources from IaaS clouds cheaply, yet execute the users' jobs in time. To achieve this, cloud brokers must use scheduling policies that match diverse requirements. Finding scheduling polices that can schedule diverse workloads with zero waiting time yet cheaply is the focus of this paper.