在复杂业务系统中，多数据源切换已成为必备技能。本文将深入剖析三种主流实现方案，带你从入门到精通！

一、多数据源应用场景

读写分离：主库负责写操作，从库处理读请求
多租户系统：不同租户使用独立数据库
分库分表：业务数据按规则分散存储
多数据库类型：同时使用MySQL、Oracle等异构数据库

二、3种实现方案对比

方案	实现复杂度	侵入性	维护成本	适用场景
AbstractRoutingDataSource	中等	高	高	简单读写分离
多SqlSessionFactory	高	高	高	异构数据库
dynamic-datasource	低	低	低	复杂多数据源

三、方案一：AbstractRoutingDataSource

实现原理

通过继承Spring的AbstractRoutingDataSource类，动态路由数据源

实现步骤

1. 添加依赖

<dependencies><dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-web</artifactId></dependency><dependency><groupId>org.springframework.boot</groupId><artifactId>spring-boot-starter-jdbc</artifactId></dependency><dependency><groupId>mysql</groupId><artifactId>mysql-connector-java</artifactId><version>8.0.27</version></dependency>
</dependencies>

2. 配置多数据源

# application.yml
spring:datasource:master:driver-class-name: com.mysql.cj.jdbc.Driverjdbc-url: jdbc:mysql://localhost:3306/master_dbusername: rootpassword: root123slave:driver-class-name: com.mysql.cj.jdbc.Driverjdbc-url: jdbc:mysql://localhost:3306/slave_dbusername: rootpassword: root123

3. 动态数据源配置类

@Configuration
public class DataSourceConfig {@Bean@ConfigurationProperties("spring.datasource.master")public DataSource masterDataSource() {return DataSourceBuilder.create().build();}@Bean@ConfigurationProperties("spring.datasource.slave")public DataSource slaveDataSource() {return DataSourceBuilder.create().build();}@Beanpublic DataSource dynamicDataSource() {Map<Object, Object> targetDataSources = new HashMap<>();targetDataSources.put("master", masterDataSource());targetDataSources.put("slave", slaveDataSource());DynamicDataSource dynamicDataSource = new DynamicDataSource();dynamicDataSource.setTargetDataSources(targetDataSources);dynamicDataSource.setDefaultTargetDataSource(masterDataSource());return dynamicDataSource;}
}

4. 自定义动态数据源

public class DynamicDataSource extends AbstractRoutingDataSource {@Overrideprotected Object determineCurrentLookupKey() {return DataSourceContextHolder.getDataSourceKey();}public static class DataSourceContextHolder {private static final ThreadLocal<String> contextHolder = new ThreadLocal<>();public static void setDataSourceKey(String key) {contextHolder.set(key);}public static String getDataSourceKey() {return contextHolder.get();}public static void clearDataSourceKey() {contextHolder.remove();}}
}

5. 自定义注解切换数据源

@Target({ElementType.METHOD, ElementType.TYPE})
@Retention(RetentionPolicy.RUNTIME)
@Documented
public @interface DataSource {String value() default "master";
}

6. AOP切面实现动态切换

@Aspect
@Component
public class DataSourceAspect {@Before("@annotation(dataSource)")public void beforeSwitchDataSource(JoinPoint point, DataSource dataSource) {String dataSourceKey = dataSource.value();DynamicDataSource.DataSourceContextHolder.setDataSourceKey(dataSourceKey);}@After("@annotation(dataSource)")public void afterSwitchDataSource(JoinPoint point, DataSource dataSource) {DynamicDataSource.DataSourceContextHolder.clearDataSourceKey();}
}

7. 业务层使用示例

@Service
public class UserService {@Autowiredprivate JdbcTemplate jdbcTemplate;// 使用主库@DataSource("master")public void createUser(User user) {String sql = "INSERT INTO users(name, email) VALUES(?, ?)";jdbcTemplate.update(sql, user.getName(), user.getEmail());}// 使用从库@DataSource("slave")public List<User> getAllUsers() {String sql = "SELECT * FROM users";return jdbcTemplate.query(sql, new BeanPropertyRowMapper<>(User.class));}
}

方案优缺点

优点：

纯Spring实现，无第三方依赖
灵活控制数据源切换

缺点：

事务管理复杂
需手动处理连接池
切换逻辑侵入业务代码

四、方案二：多SqlSessionFactory

实现原理

为每个数据源创建独立的MyBatis SqlSessionFactory

实现步骤

1. 主数据源配置

@Configuration
@MapperScan(basePackages = "com.example.mapper.master", sqlSessionFactoryRef = "masterSqlSessionFactory")
public class MasterDataSourceConfig {@Bean@ConfigurationProperties("spring.datasource.master")public DataSource masterDataSource() {return DataSourceBuilder.create().build();}@Beanpublic SqlSessionFactory masterSqlSessionFactory(@Qualifier("masterDataSource") DataSource dataSource) throws Exception {SqlSessionFactoryBean bean = new SqlSessionFactoryBean();bean.setDataSource(dataSource);bean.setMapperLocations(new PathMatchingResourcePatternResolver().getResources("classpath:mapper/master/*.xml"));return bean.getObject();}@Beanpublic DataSourceTransactionManager masterTransactionManager(@Qualifier("masterDataSource") DataSource dataSource) {return new DataSourceTransactionManager(dataSource);}
}

2. 从数据源配置

@Configuration
@MapperScan(basePackages = "com.example.mapper.slave", sqlSessionFactoryRef = "slaveSqlSessionFactory")
public class SlaveDataSourceConfig {@Bean@ConfigurationProperties("spring.datasource.slave")public DataSource slaveDataSource() {return DataSourceBuilder.create().build();}@Beanpublic SqlSessionFactory slaveSqlSessionFactory(@Qualifier("slaveDataSource") DataSource dataSource) throws Exception {SqlSessionFactoryBean bean = new SqlSessionFactoryBean();bean.setDataSource(dataSource);bean.setMapperLocations(new PathMatchingResourcePatternResolver().getResources("classpath:mapper/slave/*.xml"));return bean.getObject();}@Beanpublic DataSourceTransactionManager slaveTransactionManager(@Qualifier("slaveDataSource") DataSource dataSource) {return new DataSourceTransactionManager(dataSource);}
}

3. 业务层使用

@Service
public class OrderService {// 注入主库Mapper@Autowired@Qualifier("masterOrderMapper")private OrderMapper masterOrderMapper;// 注入从库Mapper@Autowired@Qualifier("slaveOrderMapper")private OrderMapper slaveOrderMapper;@Transactional(transactionManager = "masterTransactionManager")public void createOrder(Order order) {masterOrderMapper.insert(order);}@Transactional(transactionManager = "slaveTransactionManager")public Order getOrder(Long id) {return slaveOrderMapper.selectById(id);}
}

方案优缺点

优点：

各数据源完全隔离
事务管理清晰
支持异构数据库

缺点：

配置复杂，冗余代码多
Mapper需按数据源分包
动态切换不灵活

五、方案三：dynamic-datasource框架

框架优势

零侵入：通过注解实现数据源切换
功能丰富：支持读写分离、分库分表等
简单易用：简化多数据源配置

实现步骤

1. 添加依赖

<dependency><groupId>com.baomidou</groupId><artifactId>dynamic-datasource-spring-boot-starter</artifactId><version>3.5.0</version>
</dependency>

2. 配置数据源

spring:datasource:dynamic:primary: master # 默认数据源strict: false   # 是否严格匹配数据源datasource:master:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://localhost:3306/master_dbusername: rootpassword: root123slave1:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://localhost:3306/slave_db1username: rootpassword: root123slave2:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://localhost:3306/slave_db2username: rootpassword: root123oracle_db:driver-class-name: oracle.jdbc.OracleDriverurl: jdbc:oracle:thin:@localhost:1521:orclusername: systempassword: oracle123

3. 使用@DS注解切换数据源

@Service
public class ProductService {// 默认使用主库@Autowiredprivate JdbcTemplate jdbcTemplate;// 使用主库@DS("master")public void createProduct(Product product) {jdbcTemplate.update("INSERT INTO product(...) VALUES(...)");}// 随机使用从库@DS("slave")public Product getProduct(Long id) {return jdbcTemplate.queryForObject("SELECT * FROM product WHERE id = ?", new BeanPropertyRowMapper<>(Product.class), id);}// 指定特定从库@DS("slave1")public List<Product> getHotProducts() {return jdbcTemplate.query("SELECT * FROM product WHERE hot = 1", new BeanPropertyRowMapper<>(Product.class));}// 使用Oracle数据库@DS("oracle_db")public List<Category> getOracleCategories() {return jdbcTemplate.query("SELECT * FROM categories", new BeanPropertyRowMapper<>(Category.class));}
}

4. 高级功能：读写分离

spring:datasource:dynamic:primary: masterdatasource:master:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://master-host:3306/dbusername: rootpassword: root123slave_1:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://slave1-host:3306/dbusername: rootpassword: root123slave_2:driver-class-name: com.mysql.cj.jdbc.Driverurl: jdbc:mysql://slave2-host:3306/dbusername: rootpassword: root123strategy: # 读写分离配置load-balance: # 负载均衡策略slave: round_robin # 从库轮询策略

5. 事务管理

@DS("master")
@Transactional
public void placeOrder(Order order) {// 1. 扣减库存productService.reduceStock(order.getProductId(), order.getQuantity());// 2. 创建订单orderMapper.insert(order);// 3. 记录日志logService.logOrder(order);// 所有操作都在主库事务中执行
}

最佳实践技巧

数据源分组管理：

spring:datasource:dynamic:datasource:master_1: # 配置...master_2:# 配置...slave_1:# 配置...slave_2:# 配置...group:masters: master_1, master_2slaves: slave_1, slave_2

多租户数据源动态注册：

@Autowired
private DynamicRoutingDataSource routingDataSource;public void addTenantDataSource(String tenantId, DataSourceProperty property) {DataSource dataSource = dataSourceCreator.createDataSource(property);routingDataSource.addDataSource(tenantId, dataSource);
}

自定义负载均衡策略：

public class RandomStrategy implements LoadBalanceStrategy {@Overridepublic String determineDataSource(List<String> dataSourceNames) {Random random = new Random();return dataSourceNames.get(random.nextInt(dataSourceNames.size()));}
}

六、性能优化建议

连接池配置优化：

spring:datasource:dynamic:datasource:master:# ...hikari:maximum-pool-size: 20minimum-idle: 5connection-timeout: 30000idle-timeout: 600000max-lifetime: 1800000

避免频繁切换数据源：
- 将同一数据源操作集中处理
- 使用@DSTransactional管理跨库事务

监控数据源状态：

@RestController
public class DataSourceMonitor {@Autowiredprivate DynamicRoutingDataSource routingDataSource;@GetMapping("/datasources")public Map<String, DataSource> listDataSources() {return routingDataSource.getDataSources();}@GetMapping("/datasources/stats")public Map<String, Object> getDataSourceStats() {Map<String, Object> stats = new HashMap<>();routingDataSource.getDataSources().forEach((key, ds) -> {if(ds instanceof HikariDataSource) {HikariDataSource hikari = (HikariDataSource) ds;HikariPoolMXBean pool = hikari.getHikariPoolMXBean();stats.put(key, Map.of("active", pool.getActiveConnections(),"idle", pool.getIdleConnections(),"total", pool.getTotalConnections()));}});return stats;}
}

七、方案选型建议

中小型项目：优先选用dynamic-datasource，开发效率高
异构数据库系统：选择多SqlSessionFactory方案，隔离性好
需要高度定制：AbstractRoutingDataSource提供最大灵活性
云原生环境：dynamic-datasource + Seata分布式事务

八、常见问题解决方案

数据源切换失效：
- 检查方法是否被AOP代理
- 确保@DS注解在public方法上
- 避免类内部方法调用

跨库事务问题：

// 使用分布式事务
@DS("order")
@GlobalTransactional
public void createOrder(Order order) {// 操作订单库orderMapper.insert(order);// 操作库存库stockService.reduce(order.getProductId(), order.getQuantity());
}

连接泄露检测：

@Bean
public DataSource dataSource(DataSourceProperties properties) {HikariDataSource dataSource = properties.initializeDataSourceBuilder().type(HikariDataSource.class).build();dataSource.setLeakDetectionThreshold(5000); // 5秒泄露检测return dataSource;
}