Why Containerizing MySQL Is a Game‑Changer for Modern Operations

MySQL is the world’s most popular open‑source database, widely used before container and Kubernetes technologies emerged. The article shares practical experiences of MySQL containerization.

MySQL Operations Challenges

Traditional physical deployments face four major challenges: cost (hardware purchase, maintenance, upgrades, failures, data loss), manual setup for each new cluster (OS install, environment config, MySQL install, tuning, upgrades), operational scripting for various replication topologies, and lack of resource elasticity leading to high operational burden as instance counts grow.

Figure 1 illustrates these four dimensions.

Mainstream Solutions

Physical servers combined with a management platform to unify database administration and reduce operational costs.

Moving databases to the cloud (RDS services) for pay‑as‑you‑go, elastic scaling, high availability, and integrated storage.

Beyond these, containerizing databases becomes essential.

Why Containerize Databases?

According to the CNCF 2020 China Cloud‑Native Survey, over 60% of Chinese enterprises use containers in production, and 43% apply them to core business workloads.

Docker Advantages

Lightweight, standardized image creation for efficient deployment and delivery.

Lightweight virtualization (rootfs, cgroup, namespace) that reduces resource overhead.

Kubernetes as the De‑Facto Orchestrator

Built on Google Borg, Kubernetes provides a robust foundation for large‑scale distributed systems, offering operational capabilities such as routing, horizontal scaling, monitoring, backup, and disaster recovery, along with a declarative API for describing containerized workloads.

RadonDB MySQL

RadonDB MySQL is an open‑source, high‑availability, cloud‑native MySQL solution supporting a primary‑replica architecture with features like automatic backup, monitoring, auto‑scaling, and secure access. It is used in production by banks, insurance companies, and large enterprises.

Architecture Overview

Helm Edition

The Helm chart packages Kubernetes resources, simplifying deployment, configuration separation, lifecycle management, upgrades, and resource deletion.

MySQL high availability with leader election and sub‑second failover.

Cluster management, monitoring alerts, log management, and account control.

KubeSphere Application Management

Deployment can be performed via command‑line tools, as shown below.

$ xenoncli cluster status
+------------------------------------------------------+-------------------------------+--------+---------+--------------------------+---------------------+----------------+------------------------------------------------------+
| ID                         | Raft                         | Mysqld | Monitor | Backup                  | Mysql               | IO/SQL_RUNNING | MyLeader                                            |
+------------------------------------------------------+-------------------------------+--------+---------+--------------------------+---------------------+----------------+------------------------------------------------------+
| demo-radondb-mysql-0.demo-radondb-mysql.default:8801 | [ViewID:1 EpochID:2]@LEADER | UNKNOW | OFF     | state:[NONE]            | [ALIVE] [READWRITE] | true/true      | demo-radondb-mysql-0.demo-radondb-mysql.default:8801 |
+------------------------------------------------------+-------------------------------+--------+---------+--------------------------+---------------------+----------------+------------------------------------------------------+

Operator Edition

The Operator version provides stateful service automation for complex applications, extending Helm capabilities by:

Listening to the Kubernetes API to handle instance lifecycle events and ensure data continuity.

Supporting cross‑region disaster recovery, fixed IPs, and node placement.

Automatically detecting and repairing replication anomalies or latency issues.

Future features include node scaling, automated upgrades, backup & restore, fault‑tolerant failover, automatic node rebuild, service restart, API‑driven account management, online migration, and more.