What Is the New Chinese DevOps Standard and How Does It Shape Modern Operations?

1. DevOps Standard System

On November 17, 2017, the Cloud Computing Open Source Industry Alliance co‑hosted the first Gold Medal Operations Summit in Shanghai with the Efficient Operations Community. Guided by the Ministry of Industry and Information Technology, the alliance, led by the China Academy of Information and Communications Technology (CAICT), released two completed standards (DevOps Standard and Blue Whale Operations Standard). This article focuses on the DevOps Standard.

1.1 DevOps Framework

The DevOps standard released is a beta version, an experiential edition in industry terminology.

Why create a DevOps standard when many best practices and open‑source tools already exist? The term “DevOps” is often used inconsistently—some equate it with automation, others with developers performing operations, or vendors describing it from a product perspective. This standard aims to clarify the concept, framework, and capability maturity.

First “正” (concept): the definition and scope of DevOps.

Second “正” (framework): the processes, organization, and tools that constitute DevOps.

Third “正” (capability): a maturity model that lets teams identify their current level and the steps to advance.

With these three “正”, the standard provides practical guidance for operations staff, CIOs, and others to align processes, structures, and implementation steps.

1.2 DevOps Standard Series – First Three Parts

The standard consists of seven parts. Part 1 is the overall architecture, covering concept, framework, and capability grading. Parts 2‑4 detail the maturity models for agile development, continuous delivery, and technical operations. Part 5 addresses application architecture, Part 6 security management, and Part 7 organizational structure.

The “Research‑Operations Integrated Capability Maturity Model” is the first DevOps standard worldwide.

The standard serves two main purposes:

Clarify DevOps concepts, framework, and required capability levels.

Provide concrete guidance for processes, organization, and implementation so enterprises can improve their DevOps capabilities.

The released draft covers the first three parts.

2. Detailed Explanation of the First Three Parts

2.1 Part 1: Overall Architecture

The Research‑Operations Integrated (DevOps) capability maturity model spans the entire software delivery lifecycle. It is divided into three domains: processes (agile development management, continuous delivery, technical operations), application architecture, and organizational structure.

Process domain : subdivided into agile development management, continuous delivery, and technical operation, each with its own maturity criteria.

Application architecture : concerns development, testing, deployment, and the use of micro‑services or other architectures.

Security management : addresses security considerations throughout the end‑to‑end DevOps flow.

Organizational structure : examines how different departments (development, operations, etc.) must adapt to support DevOps.

2.2 Part 2: Agile Development Process

The agile development process includes demand management, iteration planning, and agile process management, each evaluated on five dimensions (people, mechanisms, tools, etc.).

Demand collection : defines single demand points, full demand view, and backlog creation.

Demand analysis : refines requirements into user stories and prioritizes them by business value.

Demand‑case management : links user‑story acceptance criteria with test cases.

Demand acceptance : emphasizes frequent, end‑to‑end functional verification and rapid feedback.

2.3 Part 3: Continuous Delivery Process

Continuous delivery is broken into several sub‑processes, each with maturity dimensions:

Configuration management : version control and traceability.

Build and continuous integration : build practices and CI pipelines.

Test management : test strategy, code quality, and automation.

Deployment and release management : deployment modes and continuous deployment pipelines.

Environment management : lifecycle, consistency, and versioning of environments.

Data management : test data and change management.

Metrics and feedback : measurement indicators and improvement loops.

These components together form a comprehensive DevOps standard that guides enterprises in assessing and advancing their capabilities.