Abstract:

Cloud-native applications require reliable and efficient deployment mechanisms to ensure scalability, availability, and operational consistency. Kubernetes has become the leading container orchestration platform, while Helm simplifies application deployment through package-based management. However, conventional Helm-based deployment approaches primarily focus on deployment automation and provide limited support for intelligent validation, real-time resource monitoring, automated rollback, and integrated performance analysis. To address these limitations, this paper proposes an Intelligent Kubernetes Helm Automation Framework for cloud-native application deployment. The framework integrates automated Helm chart validation, intelligent deployment execution, resource monitoring, automatic rollback, and performance analysis into a unified deployment workflow. The proposed framework was implemented using Python, Kubernetes, and Helm, and evaluated through 50 deployment experiments under a controlled environment. Experimental results demonstrate that the framework successfully completed 49 deployments, achieving a 98% deployment success rate, while only one deployment required automatic rollback. The average deployment time was 51.91 seconds, with average CPU, memory, and disk utilization of 56.89%, 60.88%, and 48.89%, respectively. In addition, the framework achieved 20.35% resource optimization, while the intelligent rollback mechanism reduced the average rollback and recovery times to 0.80 seconds and 1.03 seconds, respectively. These results demonstrate that the proposed framework enhances deployment reliability, resource efficiency, and failure recovery, making it a practical solution for intelligent Kubernetes-based cloud-native application deployment.