Spring Boot集成sharding-jdbc實現分庫分表

一、水平分割

1、水平分庫

1)、概念：
以字段為依據，按照一定策略，將一個庫中的數據拆分到多個庫中。
2)、結果
每個庫的結構都一樣；數據都不一樣；
所有庫的並集是全量數據；

2、水平分表

1)、概念
以字段為依據，按照一定策略，將一個表中的數據拆分到多個表中。
2)、結果
每個表的結構都一樣；數據都不一樣；
所有表的並集是全量數據；

二、Shard-jdbc 中間件

1、架構圖

 

2、特點

1)、Sharding-JDBC直接封裝JDBC API，舊代碼遷移成本幾乎為零。
2)、適用於任何基於Java的ORM框架，如Hibernate、Mybatis等 。
3)、可基於任何第三方的數據庫連接池，如DBCP、C3P0、 BoneCP、Druid等。
4)、以jar包形式提供服務，無proxy代理層，無需額外部署，無其他依賴。
5)、分片策略靈活，可支持等號、between、in等多維度分片，也可支持多分片鍵。
6)、SQL解析功能完善，支持聚合、分組、排序、limit、or等查詢。

 

三、項目演示

 

 

核心代碼塊

數據源配置文件

spring:
  datasource:
    # 數據源：shard_one
    dataOne:
      type: com.alibaba.druid.pool.DruidDataSource
      druid:
        driverClassName: com.mysql.jdbc.Driver
        url: jdbc:mysql://localhost:3306/shard_one?useUnicode=true&characterEncoding=UTF8&zeroDateTimeBehavior=convertToNull&useSSL=false username: root password: 123 initial-size: 10 max-active: 100 min-idle: 10 max-wait: 60000 pool-prepared-statements: true max-pool-prepared-statement-per-connection-size: 20 time-between-eviction-runs-millis: 60000 min-evictable-idle-time-millis: 300000 max-evictable-idle-time-millis: 60000 validation-query: SELECT 1 FROM DUAL # validation-query-timeout: 5000 test-on-borrow: false test-on-return: false test-while-idle: true connectionProperties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000 # 數據源：shard_two dataTwo: type: com.alibaba.druid.pool.DruidDataSource druid: driverClassName: com.mysql.jdbc.Driver url: jdbc:mysql://localhost:3306/shard_two?useUnicode=true&characterEncoding=UTF8&zeroDateTimeBehavior=convertToNull&useSSL=false username: root password: 123 initial-size: 10 max-active: 100 min-idle: 10 max-wait: 60000 pool-prepared-statements: true max-pool-prepared-statement-per-connection-size: 20 time-between-eviction-runs-millis: 60000 min-evictable-idle-time-millis: 300000 max-evictable-idle-time-millis: 60000 validation-query: SELECT 1 FROM DUAL # validation-query-timeout: 5000 test-on-borrow: false test-on-return: false test-while-idle: true connectionProperties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000 # 數據源：shard_three dataThree: type: com.alibaba.druid.pool.DruidDataSource druid: driverClassName: com.mysql.jdbc.Driver url: jdbc:mysql://localhost:3306/shard_three?useUnicode=true&characterEncoding=UTF8&zeroDateTimeBehavior=convertToNull&useSSL=false username: root password: 123 initial-size: 10 max-active: 100 min-idle: 10 max-wait: 60000 pool-prepared-statements: true max-pool-prepared-statement-per-connection-size: 20 time-between-eviction-runs-millis: 60000 min-evictable-idle-time-millis: 300000 max-evictable-idle-time-millis: 60000 validation-query: SELECT 1 FROM DUAL # validation-query-timeout: 5000 test-on-borrow: false test-on-return: false test-while-idle: true connectionProperties: druid.stat.mergeSql=true;druid.stat.slowSqlMillis=5000

數據庫分庫策略

/** * 數據庫映射計算 */ public class DataSourceAlg implements PreciseShardingAlgorithm<String> { private static Logger LOG = LoggerFactory.getLogger(DataSourceAlg.class); @Override public String doSharding(Collection<String> names, PreciseShardingValue<String> value) { LOG.debug("分庫算法參數 {},{}",names,value); int hash = HashUtil.rsHash(String.valueOf(value.getValue())); return "ds_" + ((hash % 2) + 2) ; } }
數據表1分表策略

/** * 分表算法 */ public class TableOneAlg implements PreciseShardingAlgorithm<String> { private static Logger LOG = LoggerFactory.getLogger(TableOneAlg.class); /** * 該表每個庫分5張表 */ @Override public String doSharding(Collection<String> names, PreciseShardingValue<String> value) { LOG.debug("分表算法參數 {},{}",names,value); int hash = HashUtil.rsHash(String.valueOf(value.getValue())); return "table_one_" + (hash % 5+1); } }

數據表2分表策略

/** * 分表算法 */ public class TableTwoAlg implements PreciseShardingAlgorithm<String> { private static Logger LOG = LoggerFactory.getLogger(TableTwoAlg.class); /** * 該表每個庫分5張表 */ @Override public String doSharding(Collection<String> names, PreciseShardingValue<String> value) { LOG.debug("分表算法參數 {},{}",names,value); int hash = HashUtil.rsHash(String.valueOf(value.getValue())); return "table_two_" + (hash % 5+1); } }

數據源集成配置

/** * 數據庫分庫分表配置 */ @Configuration public class ShardJdbcConfig { // 省略了 druid 配置，源碼中有 /** * Shard-JDBC 分庫配置 */ @Bean public DataSource dataSource (@Autowired DruidDataSource dataOneSource, @Autowired DruidDataSource dataTwoSource, @Autowired DruidDataSource dataThreeSource) throws Exception { ShardingRuleConfiguration shardJdbcConfig = new ShardingRuleConfiguration(); shardJdbcConfig.getTableRuleConfigs().add(getTableRule01()); shardJdbcConfig.getTableRuleConfigs().add(getTableRule02()); shardJdbcConfig.setDefaultDataSourceName("ds_0"); Map<String,DataSource> dataMap = new LinkedHashMap<>() ; dataMap.put("ds_0",dataOneSource) ; dataMap.put("ds_2",dataTwoSource) ; dataMap.put("ds_3",dataThreeSource) ; Properties prop = new Properties(); return ShardingDataSourceFactory.createDataSource(dataMap, shardJdbcConfig, new HashMap<>(), prop); } /** * Shard-JDBC 分表配置 */ private static TableRuleConfiguration getTableRule01() { TableRuleConfiguration result = new TableRuleConfiguration(); result.setLogicTable("table_one"); result.setActualDataNodes("ds_${2..3}.table_one_${1..5}"); result.setDatabaseShardingStrategyConfig(new StandardShardingStrategyConfiguration("phone", new DataSourceAlg())); result.setTableShardingStrategyConfig(new StandardShardingStrategyConfiguration("phone", new TableOneAlg())); return result; } private static TableRuleConfiguration getTableRule02() { TableRuleConfiguration result = new TableRuleConfiguration(); result.setLogicTable("table_two"); result.setActualDataNodes("ds_${2..3}.table_two_${1..5}"); result.setDatabaseShardingStrategyConfig(new StandardShardingStrategyConfiguration("phone", new DataSourceAlg())); result.setTableShardingStrategyConfig(new StandardShardingStrategyConfiguration("phone", new TableTwoAlg())); return result; } }