A Modified Bat Mechanism for Virtual Machine Migration in a Cloud Environment

A Modified Bat Mechanism for Virtual Machine Migration in a Cloud Environment

연구 분야: Cryptography

학회: SN Computer Science

In cloud computing, virtual machines (VMs) must be moved from one overloaded server to another to control impending resource requests and reduce the performance overhead. Hence, the main objective is to carry out the VM migration as effectively as possible by shortening the migration process, lowering the Makespan, and boosting the throughput of the VMs. Here, proposed a modified Bat mechanism that addresses the aforementioned problems and completes the VM migration effectively. The fitness function calculation strategy of spider monkey optimization is added to the standard Bat algorithm inside this proposed mechanism to improve each Bat’s fitness calculation and provide more opportunities to determine the fittest move quickly. The best move enables VMs to connect with the closest PMs (physical machines) and complete migration promptly rather than being stuck. This proposed mechanism overcomes the issue of the standard Bat algorithm that gets stuck in local optima due to inadequate exploration. The fitness function, migration time, makespan, and throughput performance metrics have been used to calculate the effectiveness of the proposed algorithm as compared to the Standard Bat algorithm, PSO (particle swarm optimization), and CS (cuckoo search) algorithm by using the CloudSimPlus simulator. The migration time of the proposed mechanism is 25% faster than that of the standard Bat algorithm, 27% more than PSO, and 35% more than the CS algorithm. It also outperforms other traditional algorithms in makespan, fitness, and throughput.