Analysis Of Availability Using Virtual Machine Placement Algorithms For Cloud Iaas

Cloud Computing paradigm is most popular for its flexibility in resourcernprovisioning by creating virtual machines of the requested specification on thernunderlying physical infrastructure. Infrastructure-as-a-Service, which requires arnprovider to allocate virtual machines (VMs), has significant impact on thernavailability of accommodated applications. Due to this, several algorithms havernbeen proposed for the VM placement problem with little or no objective comparisonrnhiding the fact which one works best or what factors influence the availability of thernalgorithms. In this thesis work, comparison of four algorithms using metrics ofrnavailability is presented. The findings showed that the impact of VM placementrnalgorithms along with memory and CPU utilization play an important role inrnmaintaining the availability of IaaS for cloud. As a result, algorithms which performrnwell on one metric perform poorly on the other metrics underlining the importancernof objectively comparing the availability of competing algorithms and highlightingrnthe importance of thorough empirical studies not only for power consumption andrnother QoS attributes but also for availability as one important aspect of cloudrncomputing.

Get Full Work

Report copyright infringement or plagiarism