Understanding the DevOps Toolchain: SCM, Automation, and Cloud

DevOps covers a very wide range of practices; its knowledge base can be split into two parts: methodology and toolchain. While the methodology derives from agile principles, this article focuses on the toolchain side.

Recently a DevOps toolchain periodic table was shared; the author likens it to the chemistry periodic table and admits it feels overwhelming at first.

Because humans struggle to remember complex and scattered data, the author simplifies the toolchain into a single diagram that highlights three core infrastructure layers needed to implement a DevOps toolchain:

SCM (Source Code Management) system

Automation system

Cloud (or any scalable, self‑service, virtualized environment)

SCM – Configuration Management System

Configuration management is the lowest‑level foundation of DevOps. Whether expressed as Configuration‑as‑Code or Infrastructure‑as‑Code, it treats environments as code, versioning them to enable rapid and repeatable provisioning.

Typical SCM tools include Git, SVN, Mercurial, GitHub, Bitbucket, etc. The choice of SCM influences how well subsequent automation and cloud tools can integrate; Git is currently the most popular and well‑supported option.

SCM stores two layers of code:

AppCode – the application source code

EnvCode – environment‑related code, which can be further split into configuration (static settings) and configuration data (dynamic values such as config files or environment URLs).

Automation – Automation System

Automation is essential in DevOps and is usually driven by build systems such as Jenkins, TeamCity, Travis CI, etc. However, build systems alone are insufficient; a full pipeline also needs testing, dependency management, container building, packaging, and orchestration tools like JUnit, Selenium, NuGet, NPM, JMeter, and many others.

All these automation artifacts eventually converge on a node called MOF, defined by the DMTF standard. Although DMTF has published the standard for years, support in mainstream tools like Puppet and Chef remains limited, leading the author to remain skeptical about its widespread adoption.

Cloud – Cloud Computing

Virtualization and cloud computing provide the elasticity and self‑service capabilities that make DevOps practical; without them, many DevOps ideas would remain theoretical.

Implementing DevOps requires understanding the APIs offered by various cloud platforms, as both automation systems and SCM outputs ultimately invoke these APIs to deploy applications.

The author wrote this article to organize the abundant DevOps information and hopes it helps readers locate a starting point for building their own knowledge system.

Please follow the WeChat public account devopshub for more DevOps integration content.