DevOps and Database CI/CD

The incorporation of databases into DevOps practices and Continuous Integration/Continuous Deployment (CI/CD) pipelines has been a long-standing problem and a topic of ongoing research from 2015 to 2024. Previously, databases were typically separated from the automation and agility provided by DevOps practices, leading to deployments that were infrequent, data integrity problems, and extended downtimes during database migration processes. Although initial research was centered on automating database schema migrations, there was still a huge gap in effectively managing database changes in multi-environment and cloud-native applications, particularly for large applications, microservices, and serverless systems. Past work has filled these gaps by providing solutions like automated migration tools (e.g., Liquibase, Flyway), version-controlled database schema management, and continuous database testing platforms. Yet issues like application code and database schema compatibility, handling gigantic migrations, and high availability during deployment are still common. Work has also begun exploring AI-based solutions to anticipate deployment problems, and predictive rollback platforms to reduce downtime. Even with these developments, an integrated framework that adequately addresses the intricacies that come with multi-cloud and hybrid-cloud environments, particularly in the application of machine learning to real-time database migration management in CI/CD pipelines, does not exist. This paper consolidates the work of the past decade, determines the existing gaps, and indicates possible avenues for the development of database CI/CD practices to address the needs of contemporary distributed systems. The study demands additional research on cross-cloud synchronization tools, serverless database automation, and AI-based predictive methods for enhancing deployment