Custom Resource-Aware Scheduling, Dynamic Scaling and Selective Communication in Docker Desktop

In the existing scenario, the default scheduling mechanism in Kubernetes often limits the flexibility required to address specific resource and operational needs on Docker Desktop. To address this, a customized resource-aware scheduling and dynamic scaling solution is developed, enabling enhanced control over workload distribution and resource allocation. This solution empowers application developers to replace the default Kubernetes scheduler, offering a tailored approach to managing workloads effectively based on specific requirements.The proposed framework integrates node affinity properties for optimized server selection, a metaheuristic simulated annealing algorithm for configuration optimization, and selective communication between Docker containers and external resources using AWS services. Additionally, a Binary Search Decision Strategy (BSDS) is proposed to identify optimal nodes for scaling while minimizing search time. These methodologies collectively improve resource utilization and operational efficiency, providing a robust alternative to default Kubernetes-based scheduling.