Model Driven Architecture for Continuous Software Quality Monitoring in Cloud Native Systems

Abstract

Cloud-native systems require continuous quality monitoring because services, containers, and deployment states change rapidly during operation. This study presents a model-driven architecture for continuous software quality monitoring in cloud-native environments. The framework combines service architecture models, deployment models, observability models, quality constraint models, and runtime adaptation models within a synchronized monitoring process. Live metrics, logs, traces, and orchestration events are mapped to system models and used to calculate a continuous quality score based on response time, availability, reliability, error rate, resource utilization, and service stability. The framework was evaluated during deployment, load variation, service degradation, and runtime adaptation. Results showed that the proposed architecture detected quality decline earlier than rule-based monitoring and static dashboard assessment. It also tracked recovery more accurately after scaling, workload redistribution, and container restart actions. The framework supports early warning, adaptive control, and informed operational decisions in dynamic cloud-native software systems while reducing manual monitoring effort.