Planning DevOps Infrastructure for Traditional Enterprises: A Comprehensive Capability Blueprint

DevOps originated in internet companies such as Netflix and AWS, where hundreds of deployments occur daily; traditional enterprises now seek similar speed and efficiency but often lack a clear plan for the required infrastructure and capabilities.

To build a suitable DevOps foundation, the article examines five dimensions—foundation, development, testing, operations, and project management—identifying essential capabilities for each and mapping them to popular industry tools, while emphasizing that these mappings are illustrative rather than prescriptive.

Foundation : A robust cloud computing capability is the cornerstone, requiring self‑service and environment orchestration (IaaS, e.g., VMWare), elastic scaling (PaaS, e.g., Kubernetes), and SaaS‑based services. Private or hybrid clouds are often preferred for security, though public clouds are also viable.

Development : Focuses on development efficiency, code quality, and real‑time feedback. Key abilities include distributed source control (GitLab), continuous integration and deployment (Jenkins), dependency management (Nexus), static code analysis (Sonar/Fortify/InSpec), unit testing, test artifact management, visual dashboards, and email notifications.

Testing : Emphasizes test efficiency and real‑time feedback. Required abilities are automated testing (Jenkins), parallel test execution (Selenium‑Grid), visual result displays, and email notifications. Practices such as integrating automated tests into the delivery pipeline and using tools like JMeter or Cucumber are suggested.

Operations : Centers on change‑risk control, real‑time operational feedback, and production incident response. Capabilities include production change management (Service Desk), artifact management (Nexus), automated production deployment (CodeDeploy), monitoring (Prometheus), visualization (Grafana or TV walls), mobile alerts, incident management, and knowledge sharing (Confluence). The article notes the tension between traditional ITSM processes and DevOps agility.

Project Management : Requires iteration support, analytics, change tracking, and real‑time feedback. Abilities include backlog and user‑story management (Jira), data reporting (Jira), linking requirements, code changes, CI pipelines, and documentation (Jira/Confluence), plus visual dashboards.

The article then presents a panoramic view of the DevOps infrastructure architecture, showing the cloud platform as the left side of a triangle and the capability/tool layer on the right, stressing that the architecture must evolve with business needs.

Finally, it discusses the trade‑offs of a centralized one‑stop DevOps platform, highlighting issues of flexibility and operational support, and cites Huawei’s CodeArts as a relatively complete solution that balances comprehensive capabilities with extensibility through APIs and integration points.