Service-Aware Cloud-to-Cloud Migration of Multiple Virtual Machines

Abstract

Cloud-to-Cloud (C2C) migration enables the organizations to switch among various cloud environments without rebuilding the whole system on another cloud. Numerous virtual machines (VMs) placed in one cloud-computing environment during the C2C migration are moved to another. Service running on the multiple VMs becomes unavailable during the C2C migration because the conventional migration schemes stop the VMs for a while. The migration cost due to the increased downtime of the service deployed on the VMs may be increased by inappropriate C2C migration operation. We propose a service-aware strategy for C2C migration of services on multiple VMs, which analyzes the dependency of multiple VMs, using network traffic intensity to determine the migration sequence of dependent VMs in order to decrease the service downtime. We implement the proposed migration method in an OpenStackbased testbed environment. Our experimental results illustrate that the dependency among the VMs is successfully identified for C2C migration. The proposed method that exploits the dependency among the VMs significantly reduces the service downtime, while the service downtime increases exponentially, when the VMs are migrated randomly. In comparison with the random VM migration, the proposed scheme reduced the average service downtime by over 50% during the C2C migration.