Min-max exclusive virtual machine placement in cloud computing for scientific data environment

Abstract In cloud computing, there is a trade-off between SLAV (Service Level Agreement Violation) and system operating cost. Violation rates can be decreased when using more hosts, which increases costs. Therefore, to manage the resources of those hosts efficiently, finding an optimal balancing point cost critical. addition, cost-effective load approach based on proper migration VMs (Virtual Machines) in needed. For this purpose, some indicators are also necessary identify abnormal that violate SLA. One primary indicators, CPU usage, closely related energy consumption used reduce effectively. Our focuses special environment such as for scientific data. Here, most jobs data-intensive large amount disk operations required. Owing likely affect performance degradation host, bandwidth usage well should considered. study, we propose Min-Max Exclusive VM Placement (MMEVMP) strategy minimize both consumption. The current working called KIAF (KISTI Analysis Facility), CERN ALICE experimental data analysis, analyze characteristics within it. experiment, lightweight version CloudSim simulator was developed results were compared other methods different policies. evaluation showed our proposed SLA violation reasonably cost-effectively.