Multiple-Backup Resource Allocation for Virtual Machines with Probabilistic Protection in Cloud

The paper proposes a model to minimize backup capacity while limiting the probability of unsuccessful primary resource backup during multiple physical machine (PM) failures. In this model, when a PM containing virtual machines (VMs) fails, VMs in that PM can be recovered by available backup resources in backup PMs. By setting the priority of protecting VMs and backup PMs that each VM is protected by, we obtain the probability of unsuccessful backup. We introduce heuristic approaches to allocate backup capacity with the least required capacity. The proposed model can reduce the total required backup capacity compared to the conventional models.