Master ShardingJDBC: Simplify Database Sharding, Read/Write Splitting & Transactions

ShardingJDBC Overview

ShardingJDBC (Apache ShardingSphere JDBC) is a relational database middleware that simplifies operations on sharded and read‑write separated databases.

Key Features

Zero migration cost by directly wrapping the JDBC API.

Compatible with any Java ORM such as Hibernate or MyBatis.

Works with any third‑party connection pool (DBCP, C3P0, BoneCP, Druid, etc.).

Provided as a jar without a proxy layer; no extra deployment required.

Flexible sharding strategies supporting equality, BETWEEN, IN, multi‑key sharding.

Comprehensive SQL parsing (aggregation, grouping, ordering, LIMIT, OR, etc.).

Core Functions

The main functions are data sharding and read‑write separation, allowing applications to access multiple data sources transparently.

Data Sharding Capabilities

Database and table partitioning.

Read‑write separation.

Customizable sharding strategies.

Distributed primary key generation using Snowflake algorithm.

Distributed Transaction Support

Standardized transaction API.

Two‑phase commit.

Flexible (soft) transactions.

Database Governance

Dynamic configuration.

Orchestration and governance.

Data masking.

Visual traceability.

Architecture Diagram

Practical Example

The following Spring Boot project demonstrates read‑write separation using MyBatis, Druid, and ShardingJDBC with MySQL.

pom.xml Dependencies

Include spring-boot-starter-web, mybatis-spring-boot-starter, druid-spring-boot-starter, and sharding-jdbc-spring-boot-starter.

application.yml Configuration

Configure multiple data sources, master‑slave load‑balance algorithm (round_robin or random), and other ShardingJDBC settings.

Sample Controller

@RestController
public class UserController {
    @Autowired
    private UserService userService;

    @GetMapping("/users")
    public Object list() {
        return userService.list();
    }

    @GetMapping("/add")
    public Object add(@RequestParam Integer id,
                      @RequestParam String username,
                      @RequestParam String password) {
        User user = new User();
        user.setId(id);
        user.setUsername(username);
        user.setPassword(password);
        return userService.addUser(user);
    }
}

Verification

After invoking the two APIs, insert logs appear in the master database directory and query logs appear in the slave directory, confirming that ShardingJDBC has achieved read‑write separation.

Query SELECT u.* FROM user u