6G DevOps: Facilitating AI and Yielding Systematic Insights in 6G Evolution

The increasing complexity of 6G networks, driven by advanced techniques such as virtualization, AI/ML integration, and disaggregated architectures, creates unprecedented challenges for testing and validating network functions before deployment. Traditional testing approaches become inadequate when dealing with software-based Virtualized Network Functions (VNFs) that require rapid iteration and real-world performance validation. This paper introduces the concept of 6G DevOps, with Continuous Testing (CT) as the cornerstone of the CI/CD 2 /CT pipeline, addressing the critical need for automated, scalable, and repeatable VNF testing methodologies. We propose a comprehensive CT framework specifically designed for 6G VNFs, integrating automation tools, real-world network emulation, and performance measurement to enable systematic validation and AI/ML model training. A case study on RAN performance profiling is presented to demonstrate how 6G DevOps provides critical insights into VNF resource utilization and performance, facilitating efficient network operation and AI/ML model training. The proposed methodology ensures agility, reliability, and efficiency in 6G network evolution, making it an indispensable framework for network operators and device vendors.