CODEF: Dynamic Cloud\u2013Edge Resource Demand Dataset from Kubernetes Experiments and Stress Testing

This dataset contains real-world CPU and memory utilization time-series collected from controlled cloud\u2013edge experiments on Kubernetes-based infrastructures. The data were generated using the CODEF experimentation framework across multiple Kubernetes distributions (Vanilla Kubernetes, K3s) and Container Network Interfaces (Flannel, Calico) under dynamic and heterogeneous workload conditions. The experiments include pod-scaling scenarios, compute-intensive stress workloads, and multi-phase workload transitions, reflecting realistic edge-service behavior. Metrics were gathered over approximately 280 hours of experimentation. The dataset supports research on machine learning\u2013driven network and service management, resource prediction, anomaly detection, and scalable edge-cloud orchestration under resource constraints.