Methods for Evaluating the Reliability of Web Portals at Different Stages of Development Using the Integration of Immutable Infrastructure and Containerization

Summary. The article proposes a methodology for evaluating the dependability of web portals at various stages of their life cycle through the integration of Immutable Infrastructure and containerization technologies. As web systems grow in complexity and functional load, and as the demand for high availability and fault tolerance increases, traditional reliability as- sessment approaches based on defect density, error frequency, and test coverage become insufficient. The study substantiates the relevance of combining classical and modern relia- bility metrics within DevOps practices and CI/CD pipelines.
Immutable Infrastructure entails the full replacement of system components during updates, eliminating configuration drift and ensuring environmental consistency. Contain- erization isolates software components, enhances execution repeatability, simplifies scalability, and improves recovery processes. Together, these technologies form the foundation for stable and predictable web portal performance under diverse operational conditions.
The research systematizes key reliability indicators, including Mean Time Between Failures (MTBF), Mean Time To Recovery (MTTR), downtime duration, error frequency, defect density, test coverage, and cyclomatic complexity. The relevance of each metric is defined with respect to the corresponding development stage, from architectural design to deployment.
The study also analyzes data collection and interpretation tools such as version control systems, automated testing frameworks, monitoring solutions, and orchestration platforms like Kubernetes and Docker. The use of SonarQube, Prometheus, and Terraform is examined in the context of automating reliability metric tracking and early risk detection.
It is demonstrated that the proposed approach reduces recovery time by up to 15%, decreases error frequency by up to 20%, and enhances overall environment stability. The findings offer practical value to developers, testers, and DevOps engineers seeking to improve system dependability in compliance with ISO 4.2.5.2 (availability) and 4.2.5.4 (recoverability) standards. This scientific article is devoted to the development of a methodology for evaluating the reliability of web portals at different stages of their life cycle through the integration of immutable infrastructure and containerization technologies. With the growing complexity of web systems, increased functional load, and the need for high availability and fault tolerance, traditional approaches to reliability assessment—based primarily on test coverage, error rate, and defect density—are no longer sufficient. The article addresses this gap by proposing a structured framework for applying classical and modern reliability metrics in conjunction with DevOps and CI/CD practices.