Java实战:教你如何进行数据库分库分表
摘要:本文通过实际案例,说明如何按日期来对订单数据进行水平分库和分表,实现数据的分布式查询和操作。
本文分享自华为云社区《数据库分库分表Java实战经验总结 丨【绽放吧!数据库】》,作者: jackwangcumt。
我们知道,当前的应用都离不开数据库,随着数据库中的数据越来越多,单表突破性能上限记录时,如MySQL单表上线估计在近千万条内,当记录数继续增长时,从性能考虑,则需要进行拆分处理。而拆分分为横向拆分和纵向拆分。一般来说,采用横向拆分较多,这样的表结构是一致的,只是不同的数据存储在不同的数据库表中。其中横向拆分也分为分库和分表。
1 示例数据库准备
为了说清楚如何用Java语言和相关框架实现业务表的分库和分表处理。这里首先用MySQL数据库中创建两个独立的数据库实例,名字为mydb和mydb2,此可演示分库操作。另外在每个数据库实例中,创建12个业务表,按年月进行数据拆分。具体的创建表脚本如下:
CREATE TABLE `t_bill_2021_1` ( `order_id` bigint(20) NOT NULL COMMENT '订单id', `user_id` int(20) NOT NULL COMMENT '用户id', `address_id` bigint(20) NOT NULL COMMENT '地址id', `status` char(1) DEFAULT NULL COMMENT '订单状态', `create_time` datetime DEFAULT NULL COMMENT '创建时间', PRIMARY KEY (`order_id`) USING BTREE ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci; CREATE TABLE `t_bill_2021_2` ( `order_id` bigint(20) NOT NULL COMMENT '订单id', `user_id` int(20) NOT NULL COMMENT '用户id', `address_id` bigint(20) NOT NULL COMMENT '地址id', `status` char(1) DEFAULT NULL COMMENT '订单状态', `create_time` datetime DEFAULT NULL COMMENT '创建时间', PRIMARY KEY (`order_id`) USING BTREE ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci; -- 省略.... CREATE TABLE `t_bill_2021_12` ( `order_id` bigint(20) NOT NULL COMMENT '订单id', `user_id` int(20) NOT NULL COMMENT '用户id', `address_id` bigint(20) NOT NULL COMMENT '地址id', `status` char(1) DEFAULT NULL COMMENT '订单状态', `create_time` datetime DEFAULT NULL COMMENT '创建时间', PRIMARY KEY (`order_id`) USING BTREE ) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci;
成功执行脚本后,在MySQL管理工具中可以看到如下的示例界面:
2 分库分表实现
在Java语言下的框架中,有众多的开源框架,其中关于分库分表的框架,可以选择Apache ShardingSphere,其官网介绍说:ShardingSphere 是一套开源的分布式数据库解决方案组成的生态圈,它由 JDBC、Proxy 和 Sidecar(规划中)这 3 款既能够独立部署,又支持混合部署配合使用的产品组成。 它们均提供标准化的数据水平扩展、分布式事务和分布式治理等功能,可适用于如 Java 同构、异构语言、云原生等各种多样化的应用场景。Apache ShardingSphere 5.x 版本开始致力于可插拔架构。 目前,数据分片、读写分离、数据加密、影子库压测等功能,以及 MySQL、PostgreSQL、SQLServer、Oracle 等 SQL 与协议的支持,均通过插件的方式织入项目。官网地址为: https://shardingsphere.apache.org/index_zh.html 。
下面的示例采用Spring Boot框架来实现,相关的库通过Maven进行管理。首先给出pom.xml配置文件的定义:
<?xml version="1.0" encoding="UTF-8"?> <project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 https://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion> <parent> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-parent</artifactId> <version>2.5.3</version> <relativePath/> <!-- lookup parent from repository --> </parent> <groupId>com.example</groupId> <artifactId>wyd</artifactId> <version>0.0.1-SNAPSHOT</version> <name>wyd</name> <description>Demo project for Spring Boot</description> <properties> <java.version>1.8</java.version> <mybatis-plus.version>3.1.1</mybatis-plus.version> <sharding-sphere.version>4.0.0-RC2</sharding-sphere.version> <shardingsphere.version>5.0.0-beta</shardingsphere.version> </properties> <dependencies> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-web</artifactId> </dependency> <dependency> <groupId>org.mybatis.spring.boot</groupId> <artifactId>mybatis-spring-boot-starter</artifactId> <version>2.0.1</version> </dependency> <dependency> <groupId>com.baomidou</groupId> <artifactId>mybatis-plus-boot-starter</artifactId> <version>${mybatis-plus.version}</version> </dependency> <dependency> <groupId>org.projectlombok</groupId> <artifactId>lombok</artifactId> <optional>true</optional> </dependency> <dependency> <groupId>joda-time</groupId> <artifactId>joda-time</artifactId> <version>2.9.8</version> </dependency> <dependency> <groupId>org.apache.shardingsphere</groupId> <artifactId>sharding-jdbc-spring-boot-starter</artifactId> <version>${sharding-sphere.version}</version> </dependency> <dependency> <groupId>org.apache.shardingsphere</groupId> <artifactId>sharding-jdbc-spring-namespace</artifactId> <version>${sharding-sphere.version}</version> </dependency> <dependency> <groupId>mysql</groupId> <artifactId>mysql-connector-java</artifactId> <scope>runtime</scope> </dependency> <dependency> <groupId>org.postgresql</groupId> <artifactId>postgresql</artifactId> <scope>runtime</scope> </dependency> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-test</artifactId> <scope>test</scope> </dependency> </dependencies> <build> <plugins> <plugin> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-maven-plugin</artifactId> </plugin> </plugins> </build> </project>
其次,给出一个实体类,它对应于上述创建的数据库表t_bill,其定义如下:
package com.example.wyd.dao; import com.baomidou.mybatisplus.annotation.TableName; import lombok.Data; import java.util.Date; @Data @TableName("t_bill") public class Bill { private Long orderId; private Integer userId; private Long addressId; private String status; private Date createTime; public void setOrderId(Long orderId) { this.orderId = orderId; } public void setUserId(Integer userId) { this.userId = userId; } public void setAddressId(Long addressId) { this.addressId = addressId; } public void setStatus(String status) { this.status = status; } public void setCreateTime(Date createTime) { this.createTime = createTime; } }
映射类BillMapper定义如下:
package com.example.wyd.mapper; import com.baomidou.mybatisplus.core.mapper.BaseMapper; import com.example.wyd.dao.Bill; public interface BillMapper extends BaseMapper<Bill> { }
服务类接口定义如下:
package com.example.wyd.service; import com.baomidou.mybatisplus.extension.service.IService; import com.example.wyd.dao.Bill; public interface BillService extends IService<Bill> { }
服务类接口的实现类定义如下:
package com.example.wyd.service; import com.baomidou.mybatisplus.extension.service.impl.ServiceImpl; import com.example.wyd.dao.Bill; import com.example.wyd.mapper.BillMapper; import org.springframework.stereotype.Service; @Service public class BillServiceImpl extends ServiceImpl<BillMapper, Bill> implements BillService { }
这里我们采用了MybatisPlus框架,它可以很方便的进行数据库相关操作,而无需过多写SQL来实现具体业务逻辑。通过上述定义,通过继承接口的方式,并提供实体类的定义,MybatisPlus框架会通过反射机制来根据数据库设置来生成SQL语句,其中包含增删改查接口,具体的实现我们并未具体定义。
下面定义一个自定义的分库算法,具体实现如下:
package com.example.wyd; import org.apache.shardingsphere.api.sharding.standard.PreciseShardingAlgorithm; import org.apache.shardingsphere.api.sharding.standard.PreciseShardingValue; import java.util.Collection; //自定义数据库分片算法 public class DBShardingAlgorithm implements PreciseShardingAlgorithm<Long> { @Override public String doSharding(Collection<String> availableTargetNames, PreciseShardingValue<Long> shardingValue) { //真实数据库节点 availableTargetNames.stream().forEach((item) -> { System.out.println("actual db:" + item); }); //逻辑表以及分片的字段名 System.out.println("logicTable:"+shardingValue.getLogicTableName()+";shardingColumn:"+ shardingValue.getColumnName()); //分片数据字段值 System.out.println("shardingColumn value:"+ shardingValue.getValue().toString()); //获取字段值 long orderId = shardingValue.getValue(); //分片索引计算 0 , 1 long db_index = orderId & (2 - 1); for (String each : availableTargetNames) { if (each.equals("ds"+db_index)) { //匹配的话,返回数据库名 return each; } } throw new IllegalArgumentException(); } }
下面给出数据的分表逻辑,这个定义稍显复杂一点,就是根据业务数据的日期字段值,根据月份落入对应的物理数据表中。实现示例代码如下:
package com.example.wyd; import org.apache.shardingsphere.api.sharding.standard.PreciseShardingAlgorithm; import org.apache.shardingsphere.api.sharding.standard.PreciseShardingValue; import java.util.Collection; import java.util.Date; //表按日期自定义分片 public class TableShardingAlgorithm implements PreciseShardingAlgorithm<Date> { @Override public String doSharding(Collection<String> availableTargetNames, PreciseShardingValue<Date> shardingValue) { //真实数据库节点 availableTargetNames.stream().forEach((item) -> { System.out.println("actual db:" + item); }); //逻辑表以及分片的字段名 System.out.println("logicTable:"+shardingValue.getLogicTableName()+";shardingColumn:"+ shardingValue.getColumnName()); //分片数据字段值 System.out.println("shardingColumn value:"+ shardingValue.getValue().toString()); //获取表名前缀 String tb_name = shardingValue.getLogicTableName() + "_"; //根据日期分表 Date date = shardingValue.getValue(); String year = String.format("%tY", date); String mon =String.valueOf(Integer.parseInt(String.format("%tm", date))); //String dat = String.format("%td", date); //也可以安装年月日来分表 // 选择表 tb_name = tb_name + year + "_" + mon; //实际的表名 System.out.println("tb_name:" + tb_name); for (String each : availableTargetNames) { //System.out.println("availableTableName:" + each); if (each.equals(tb_name)) { //返回物理表名 return each; } } throw new IllegalArgumentException(); } }
数据的分库分表可以在Spring Boot的属性配置文件中进行设(application.properties):
server.port=8080 ######################################################################################################### # 配置ds0 和ds1两个数据源 spring.shardingsphere.datasource.names = ds0,ds1 #ds0 配置 spring.shardingsphere.datasource.ds0.type = com.zaxxer.hikari.HikariDataSource spring.shardingsphere.datasource.ds0.driver-class-name = com.mysql.cj.jdbc.Driver spring.shardingsphere.datasource.ds0.jdbc-url = jdbc:mysql://127.0.0.1:3306/mydb?characterEncoding=utf8 spring.shardingsphere.datasource.ds0.username = uname spring.shardingsphere.datasource.ds0.password = pwd #ds1 配置 spring.shardingsphere.datasource.ds1.type = com.zaxxer.hikari.HikariDataSource spring.shardingsphere.datasource.ds1.driver-class-name = com.mysql.cj.jdbc.Driver spring.shardingsphere.datasource.ds1.jdbc-url = jdbc:mysql://127.0.0.1:3306/mydb2characterEncoding=utf8 spring.shardingsphere.datasource.ds1.username = uname spring.shardingsphere.datasource.ds1.password = pwd ######################################################################################################### # 默认的分库策略:id取模 spring.shardingsphere.sharding.default-database-strategy.inline.sharding-column = id spring.shardingsphere.sharding.default-database-strategy.inline.algorithm-expression = ds$->{id % 2} ######################################################################################################### spring.shardingsphere.sharding.tables.t_bill.actual-data-nodes=ds$->{0..1}.t_bill_$->{2021..2021}_$->{1..12} #数据库分片字段 spring.shardingsphere.sharding.tables.t_bill.database-strategy.standard.sharding-column=order_id #自定义数据库分片策略 spring.shardingsphere.sharding.tables.t_bill.database-strategy.standard.precise-algorithm-class-name=com.example.wyd.DBShardingAlgorithm #表分片字段 spring.shardingsphere.sharding.tables.t_bill.table-strategy.standard.sharding-column=create_time #自定义表分片策略 spring.shardingsphere.sharding.tables.t_bill.table-strategy.standard.precise-algorithm-class-name=com.example.wyd.TableShardingAlgorithm ######################################################################################################### # 使用SNOWFLAKE算法生成主键 spring.shardingsphere.sharding.tables.t_bill.key-generator.column = order_id spring.shardingsphere.sharding.tables.t_bill.key-generator.type = SNOWFLAKE spring.shardingsphere.sharding.tables.t_bill.key-generator.props.worker.id=123 ######################################################################################################### spring.shardingsphere.props.sql.show = true
最后,我们给出一个定义的Controller类型,来测试分库分表的查询和保存操作是否正确。HomeController类定义如下:
package com.example.wyd.controller; import com.baomidou.mybatisplus.core.conditions.query.QueryWrapper; import com.example.wyd.dao.Bill; import com.example.wyd.service.BillService; import org.joda.time.DateTime; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RequestParam; import org.springframework.web.bind.annotation.RestController; import java.text.ParseException; import java.text.SimpleDateFormat; import java.util.Date; import java.util.List; @RestController @RequestMapping("/api") public class HomeController { @Autowired private BillService billService; //http://localhost:8080/api/query?start=2021-02-07%2000:00:00&end=2021-03-07%2000:00:00 @RequestMapping("/query") public List<Bill> queryList(@RequestParam("start") String start, @RequestParam("end") String end) { SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); try { Date date = sdf.parse(start); Date date2 = sdf.parse(end); QueryWrapper<Bill> queryWrapper = new QueryWrapper<>(); queryWrapper.ge("create_time",date) .and(qw-> qw.le("create_time", date2)).last("limit 1,10"); List<Bill> billIPage = billService.list(queryWrapper); System.out.println(billIPage.size()); billIPage.forEach(System.out::println); return billIPage; } catch (ParseException e) { e.printStackTrace(); } return null; } //http://localhost:8080/api/save?userid=999&addressId=999&status=M&date=2021-03-07%2000:00:00 @RequestMapping("/save") public String Save(@RequestParam("userid") int userId, @RequestParam("addressId") long AddressId, @RequestParam("status") String status ,@RequestParam("date") String strDate) { String ret ="0"; SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); try { Date date = sdf.parse(strDate); Bill bill = new Bill(); bill.setUserId(userId); bill.setAddressId(AddressId); bill.setStatus(status); bill.setCreateTime(date); boolean isOk = billService.save(bill); if (isOk){ ret ="1"; } } catch (ParseException e) { e.printStackTrace(); } return ret; } }
至此,我们可以用测试类初始化一些数据,并做一些初步的数据操作测试:
package com.example.wyd; import com.baomidou.mybatisplus.core.conditions.query.QueryWrapper; import com.example.wyd.dao.Bill; import com.example.wyd.dao.Order; import com.example.wyd.service.BillService; import com.example.wyd.service.OrderService; import org.joda.time.DateTime; import org.junit.jupiter.api.Test; import org.springframework.beans.factory.annotation.Autowired; import java.text.ParseException; import java.text.SimpleDateFormat; import java.util.*; public class OrderServiceImplTest extends WydApplicationTests { @Autowired private BillService billService; @Test public void testBillSave(){ for (int i = 0 ; i< 120 ; i++){ Bill bill = new Bill(); bill.setUserId(i); bill.setAddressId((long)i); bill.setStatus("K"); bill.setCreateTime((new Date(new DateTime(2021,(i % 11)+1,7,00, 00,00,000).getMillis()))); billService.save(bill); } } @Test public void testGetByOrderId(){ long id = 626038622575374337L; //根据数据修改,无数据会报错 QueryWrapper<Bill> queryWrapper = new QueryWrapper<>(); queryWrapper.eq("order_id", id); Bill bill = billService.getOne(queryWrapper); System.out.println(bill.toString()); } @Test public void testGetByDate(){ SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); try { Date date = sdf.parse("2021-02-07 00:00:00"); QueryWrapper<Bill> queryWrapper = new QueryWrapper<>(); queryWrapper.eq("create_time",date); List<Bill> billIPage = billService.list(queryWrapper); System.out.println(billIPage.size()); System.out.println(billIPage.toString()); } catch (ParseException e) { e.printStackTrace(); } } @Test public void testGetByDate2(){ SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss"); try { Date date = sdf.parse("2021-02-07 00:00:00"); Date date2 = sdf.parse("2021-03-07 00:00:00"); QueryWrapper<Bill> queryWrapper = new QueryWrapper<>(); queryWrapper.ge("create_time",date) .and(qw-> qw.le("create_time", date2)); List<Bill> billIPage = billService.list(queryWrapper); System.out.println(billIPage.size()); billIPage.forEach(System.out::println); } catch (ParseException e) { e.printStackTrace(); } } }
执行上述测试,通过后会生成测试数据。
3 验证
打开浏览器,输入网址进行查询测试:http://localhost:8080/api/query?start=2021-02-07%2000:00:00&end=2021-03-07%2000:00:00
输入如下网址进行数据新增测试:http://localhost:8080/api/save?userid=999&addressId=999&status=M&date=2021-03-07%2000:00:00
通过跟踪分析,此数据落入如下的表中,SQL语句如下:
SELECT * FROM mydb2.t_bill_2021_3 LIMIT 0, 1000
这里还需要注意,ShardingSphere 还支持分布式事务,感兴趣的可以阅读官网相关资料进行学习。