Revolutionizing multi-cloud environment: novel optimization driven QoS autonomic resource provisioning

The businesses looking to manage a variety of workloads with agility, capacity, and cost effectiveness, multi-cloud setups have become indispensable. However, sustaining strict Quality of Service (QoS) requirements across many platforms while guaranteeing smooth access to resources presents difficulties for these systems. This paper proposes a unique approach called multi-task sand cat swarm-driven autonomic resource provisioning (MSCS-ARP) that assures quality-of-service (QoS) while enabling efficient corporate procedure performance in a multi-cloud environment. Automated computing is implemented for tracking resource use and forecasting potential needed resources, after which resources are expanded by the need, to enhance efficiency and assure flexible resource delivery. The local agent uses hierarchical clustering to categorize the tasks into Central processing unit (CPU) and input/output (I/O) heavy categories to determine the resources needed for performing the arriving tasks. Next, the assessment stage uses random forest (RF) for predicting the workload need. According to the predicting results, the containers are subsequently scaled. Eventually, the global agent makes use of the multi-task sand cat swarm optimization (MSCSO) methodology to choose appropriate containers for carrying out the categorized tasks. The suggested architecture was put into practice on the CloudSim system and corporate task records were used to assess its effectiveness. The multi-cloud environment measures for response time (0.72 s), reliability (96.42%), energy consumption (99.38 kJ) and makespan (3154 ms) were employed the models' performance. Regarding successful ARP in a multi-cloud environment, the total computations demonstrated the efficacy of the suggested MSCS-ARP technique in comparison to other methodologies.