BaOA: Intelligent and Resilient Task Scheduling in Multi-Cloud Environment

Task scheduling in multi-cloud environments is a very important problem because resources are often heterogeneous, workloads are constantly changing, and high reliability and performance are to be guaranteed. The paper will introduce a new Barber Optimization Algorithm (BaOA) to map the tasks and VM efficiently and resiliently in multi-cloud environments to reduce the failure rate of the VM and increase the success rate of tasks and the system throughput. The suggested BaOA-based scheduler is tested in the CloudSim simulation environment and assessed with the help of the realistic Google Cloud workload traces and its performance is compared with the well-known metaheuristic algorithms such as Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Particle Swarm Optimization (PSO), in a systematic manner. Extensive experimental results shows that BaOA has a significant reduction in VM failure rate of up to 28.6%, task success rate of up to 17.9% and throughput of up to 21.4% as compared to the benchmark algorithms. Such performance improvements can be explained by the balanced exploration-exploitation approach of BaOA, which provides adaptive and failure-conscious scheduling decisions in conditions of significantly changing and multi-cloud scenarios. The results validate that the suggested BaOA-based scheduling platform provides a dependable, scalable, and high-performance model of next-generation multi-cloud computing systems and has a high probability of improving service reliability and the performance of the system in general.