Operations 9 min read

How Agricultural Bank’s AIR Platform Realized DevOps at Scale

This article details the Agricultural Bank’s Distributed Application Interconnection (AIR) platform’s DevOps journey, covering small‑granularity development, comprehensive configuration management, end‑to‑end automated pipelines, advanced testing automation, metric‑driven improvements, and the successful third‑level continuous delivery assessment across multiple projects.

Efficient Ops
Efficient Ops
Efficient Ops
How Agricultural Bank’s AIR Platform Realized DevOps at Scale

DevOps continuous delivery integrates operations and development teams to shorten development cycles, enable frequent releases, reduce errors, improve engineering efficiency, and lower costs. For Agricultural Bank, DevOps is a key component of digital transformation, built through standardized processes, tool integration, data visualization, and pilot evaluations to create a delivery pipeline suited to the bank’s needs.

The Distributed Application Interconnection (AIR) platform is among the first projects to pass a third‑level DevOps pilot. Serving as the bank’s service governance, runtime interface control, and application integration foundation, AIR provides micro‑service architecture support and forms a core part of the bank’s technology middle platform. By implementing DevOps at this foundational level, AIR helps the bank broadly promote DevOps across all systems.

From the project’s start, the AIR team studied each DevOps capability, continuously strengthening project‑level abilities while validating organization‑level outputs, ultimately establishing a bank‑specific DevOps framework and culture to achieve quality and efficiency gains.

Small‑Granularity Requirements, Rapid Development

During requirement analysis, developers and testers collaborate in iteration planning, allowing testers to familiarize themselves early, design test scenarios, and prepare data. Iterations focus on the smallest producible unit, assigning responsibilities clearly, tracking delivery end‑to‑end, holding daily stand‑ups, and conducting post‑iteration reviews to continuously improve delivery efficiency.

Everything Under Configuration Management, Traceable Changes

Branch strategies enforce small‑granularity development, with feature branches created and merged within 3‑5 days. TFS automatically creates feature branches, links them to requirements, and enables daily integration. Automated builds, code quality gates, and two‑level online code reviews ensure high code quality.

Source code, configuration files, build and deployment scripts, database change scripts, and test scripts are all managed under configuration control. Artifacts are stored in a unified repository with promotion flows, automatic maturity tagging, and bidirectional traceability between artifacts and source code, including embedded test reports and build records.

Full‑Process Automated Pipeline

A comprehensive automated pipeline spans development, testing, and production environments. Continuous delivery artifacts flow through automated quality gates to production. Development pipelines automate builds and deployments, enhancing self‑testing efficiency, while test pipelines cover all interface tests and automated failure scenarios, ensuring high release quality. Automated gray‑release to external and production environments improves user experience and reduces technical risk.

Strengthening Automated Testing Capability

The in‑house ATP+RCAP testing tool enables automated test case design and execution within the pipeline, with test cases version‑controlled in TFS. Data‑driven approaches separate test data from scripts, ensuring independent execution. Swagger UI auto‑generates API documentation, guaranteeing 100% interface test coverage.

Testers continuously update knowledge bases and maintain automated test error codes, improving result analysis and reducing false positives.

Fine‑Grained Metric Data Driving Improvement

A scientific measurement system and platform are essential for DevOps. The AIR team built a metric platform addressing previous data‑management pain points, contributing to an organization‑wide measurement framework. Regular metric reviews identify issues and guide continuous improvement.

DevOps standards consist of five levels; achieving level 3 in continuous delivery is considered a domestic leading benchmark, indicating that Agricultural Bank’s evaluated projects have reached a leading level of continuous delivery capability.

Projects that passed the level 3 assessment include:

Credit Middle‑Office Project

Personal Online Banking Project

Distributed Application Interconnection (AIR) Project

Value‑Added Tax Input Management Project

Financial Mini‑Store Project

automationTestingconfiguration managementDevOpsmetricsContinuous DeliveryBanking IT
Efficient Ops
Written by

Efficient Ops

This public account is maintained by Xiaotianguo and friends, regularly publishing widely-read original technical articles. We focus on operations transformation and accompany you throughout your operations career, growing together happily.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.