Apache ShardingSphere分表的简单使用和常见问题

目录

简介

什么是 Apache ShardingSphere?

分库分表的背景

使用

pom

配置

1，application.properties配置文件

2，创建配置类

分表

验证分表

常见问题

自定义分表规则未生效

---

简介

官网：Apache ShardingSphere

版本：4.x

什么是 Apache ShardingSphere?

Apache ShardingSphere 是一款分布式的数据库生态系统，可以将任意数据库转换为分布式数据库，并通过数据分片、弹性伸缩、加密等能力对原有数据库进行增强。

分库分表的背景

传统的将数据集中存储⾄单⼀数据节点的解决⽅案，在性能、可⽤性和运维成本这三⽅⾯已经难于满⾜互联⽹的海量数据场景。

随着业务数据量的增加，原来所有的数据都是在一个数据库上的，网络IO及文件IO都集中在一个数据库上的，因此CPU、内存、文件IO、网络IO都可能会成为系统瓶颈。

当业务系统的数据容量接近或超过单台服务器的容量、QPS/TPS接近或超过单个数据库实例的处理极限等，

在数据量超过阈值的情况下，索引深度的增加也将使得磁盘访问的 IO 次数增加，进而导致查询性能的下降；

⾼并发访问请求也使得集中式数据库成为系统的最⼤瓶颈。

在传统的关系型数据库⽆法满⾜互联⽹场景需要的情况下，将数据存储⾄原⽣⽀持分布式的 NoSQL 的尝试越来越多。但 NoSQL 并不能包治百病。

此时，往往是采用垂直和水平结合的数据拆分方法，把数据服务和数据存储分布到多台数据库服务器上。

使用

pom

            <!-- sharding-jdbc -->
            <dependency>
                <groupId>org.apache.shardingsphere</groupId>
                <artifactId>sharding-jdbc-spring-boot-starter</artifactId>
                <version>4.1.1</version>
            </dependency>
            <!-- 数据源治理, 动态切换  -->
            <dependency>
                <groupId>org.apache.shardingsphere</groupId>
                <artifactId>sharding-jdbc-orchestration</artifactId>
                <version>4.1.1</version>
            </dependency>
            <!-- 分布式事务 -->
            <dependency>
                <groupId>org.apache.shardingsphere</groupId>
                <artifactId>sharding-transaction-xa-core</artifactId>
                <version>4.1.1</version>
            </dependency>

配置

配置手册 :: ShardingSphere

1，application.properties配置文件

#垂直分表策略
# 配置真实数据源
spring.shardingsphere.datasource.names=m1

# 配置第 1 个数据源,数据源
spring.shardingsphere.datasource.m1.type=com.alibaba.druid.pool.DruidDataSource
spring.shardingsphere.datasource.m1.driver-class-name=com.mysql.cj.jdbc.Driver
spring.shardingsphere.datasource.m1.url=jdbc:mysql://localhost:3306/coursedb?serverTimezone=GMT%2B8
spring.shardingsphere.datasource.m1.username=root
spring.shardingsphere.datasource.m1.password=root

# 指定表的分布情况 配置表在哪个数据库里，表名是什么。水平分表，分两个表：m1.course_1,m1.course_2
spring.shardingsphere.sharding.tables.course.actual-data-nodes=m1.course_$->{1..2}

# 指定表的主键生成策略
spring.shardingsphere.sharding.tables.course.key-generator.column=cid
spring.shardingsphere.sharding.tables.course.key-generator.type=SNOWFLAKE
#雪花算法的一个可选参数
spring.shardingsphere.sharding.tables.course.key-generator.props.worker.id=1

#使用自定义的主键生成策略
#spring.shardingsphere.sharding.tables.course.key-generator.type=MYKEY
#spring.shardingsphere.sharding.tables.course.key-generator.props.mykey-offset=88

#指定分片策略 约定cid值为偶数添加到course_1表。如果是奇数添加到course_2表。
# 选定计算的字段，分片健
spring.shardingsphere.sharding.tables.course.table-strategy.inline.sharding-column= cid
# 根据计算的字段算出对应的表名。分片算法  course_$->{cid%2+1}，2进courese1, 1进course2
spring.shardingsphere.sharding.tables.course.table-strategy.inline.algorithm-expression=course_$->{cid%2+1}

# 打开sql日志输出。
spring.shardingsphere.props.sql.show=true

spring.main.allow-bean-definition-overriding=true

2，创建配置类

本文采用这一种方式配置

MyDbConfig

package com.example.demo.config;

import java.sql.SQLException;
import java.util.HashMap;
import java.util.Map;
import java.util.Properties;

import javax.sql.DataSource;

import org.apache.ibatis.session.SqlSessionFactory;
import org.apache.shardingsphere.api.config.sharding.ShardingRuleConfiguration;
import org.apache.shardingsphere.api.config.sharding.TableRuleConfiguration;
import org.apache.shardingsphere.api.config.sharding.strategy.ComplexShardingStrategyConfiguration;
import org.apache.shardingsphere.shardingjdbc.api.ShardingDataSourceFactory;
import org.apache.shardingsphere.underlying.common.config.properties.ConfigurationPropertyKey;
import org.mybatis.spring.SqlSessionTemplate;
import org.mybatis.spring.transaction.SpringManagedTransactionFactory;
import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.context.annotation.Primary;
import org.springframework.core.env.Environment;
import org.springframework.core.io.support.PathMatchingResourcePatternResolver;
import org.springframework.jdbc.datasource.DataSourceTransactionManager;
import org.springframework.transaction.annotation.EnableTransactionManagement;

import com.alibaba.druid.pool.DruidDataSource;
import com.baomidou.mybatisplus.extension.MybatisMapWrapperFactory;
import com.baomidou.mybatisplus.extension.spring.MybatisSqlSessionFactoryBean;

/**
 * @author admin
 * @version 1.0
 * @since 2022/12/13 20:30
 */
@Configuration
@EnableTransactionManagement
public class MyDbConfig {

    /**
     * 主数据源, 默认注入
     */
    @Bean(name = "dataSource")
    @Primary
    public DataSource druidDataSource(Environment environment) {
        return createDruidDataSource(environment);
    }

    /**
     * 主数据源 分表
     */
    @Bean(name = "dataSourceSharding")
    public DataSource getShardingDataSource(@Qualifier("dataSource") DataSource dataSource, Environment environment) throws SQLException {
        ShardingRuleConfiguration shardingRuleConfig = new ShardingRuleConfiguration();

        // 分表规则
        TableRuleConfiguration userTableRuleConfiguration = new TableRuleConfiguration("user", "dataSource.user");
        ComplexShardingStrategyConfiguration userComplexShardingStrategyConfiguration = new ComplexShardingStrategyConfiguration("create_time",
            new UserComplexKeysShardingTableRule());
        userTableRuleConfiguration.setTableShardingStrategyConfig(userComplexShardingStrategyConfiguration);
        shardingRuleConfig.getTableRuleConfigs().add(userTableRuleConfiguration);

        //数据源
        Map<String, DataSource> result = new HashMap<>(1);
        result.put("dataSource", dataSource);
        Properties properties = new Properties();
        properties.put(ConfigurationPropertyKey.MAX_CONNECTIONS_SIZE_PER_QUERY.getKey(), 5);
        return ShardingDataSourceFactory.createDataSource(result, shardingRuleConfig, properties);
    }

    @Bean("mybatisMapWrapperFactory")
    public MybatisMapWrapperFactory createMybatisMapWrapperFactory() {
        return new MybatisMapWrapperFactory();
    }

    @Bean(name = "sqlSessionFactory")
    public MybatisSqlSessionFactoryBean createMybatisSqlSessionFactoryBean(@Qualifier("dataSourceSharding") DataSource dataSource,
                                                                           MybatisMapWrapperFactory mybatisMapWrapperFactory) throws Exception {
        MybatisSqlSessionFactoryBean bean = new MybatisSqlSessionFactoryBean();
        bean.setDataSource(dataSource);
        bean.setTransactionFactory(new SpringManagedTransactionFactory());
        // 扫描指定目录的xml
        bean.setMapperLocations(new PathMatchingResourcePatternResolver().getResources("classpath:mapper/**/*Mapper.xml"));
        Properties prop = new Properties();
        //转驼峰
        prop.setProperty("mapUnderscoreToCamelCase", "true");
        //允许使用自动生成主键
        prop.setProperty("useGeneratedKeys", "true");
        prop.setProperty("logPrefix", "dao.");
        bean.setConfigurationProperties(prop);
        // 扫描包
        bean.setTypeAliasesPackage("com.example.demo.dao");
        bean.setObjectWrapperFactory(mybatisMapWrapperFactory);
        return bean;
    }

    @Bean(name = "sqlSessionTemplate")
    public SqlSessionTemplate sqlSessionTemplate(@Qualifier("sqlSessionFactory") SqlSessionFactory sqlSessionFactory) throws Exception {
        return new SqlSessionTemplate(sqlSessionFactory);
    }

    @Bean("transactionManager")
    public DataSourceTransactionManager createTransactionManager(@Qualifier("dataSourceSharding") DataSource dynamicDataSource) {
        return new DataSourceTransactionManager(dynamicDataSource);
    }

    /**
     * 创建数据源
     * @param env env
     * @return DruidDataSource
     */
    private static DruidDataSource createDruidDataSource(Environment env) {
        DruidDataSource dataSource = new DruidDataSource();
        dataSource.setUrl(env.getProperty("spring.datasource.druid.url"));
        dataSource.setUsername(env.getProperty("spring.datasource.druid.username"));
        dataSource.setPassword(env.getProperty("spring.datasource.druid.password"));
        dataSource.setInitialSize(env.getProperty("spring.datasource.druid.initial-size", Integer.class));
        dataSource.setMaxActive(env.getProperty("spring.datasource.druid.max-active", Integer.class));
        dataSource.setMinIdle(env.getProperty("spring.datasource.druid.min-idle", Integer.class));
        dataSource.setMaxWait(env.getProperty("spring.datasource.druid.maxWait", Integer.class));
        dataSource.setValidationQuery(env.getProperty("spring.datasource.druid.validationQuery"));
        return dataSource;
    }
}

application.properties

spring.datasource.druid.initial-size = 10
spring.datasource.druid.min-idle = 10
spring.datasource.druid.max-active = 20
spring.datasource.druid.maxWait = 6000
spring.datasourceslave.druid.initial-size = 10
spring.datasourceslave.druid.min-idle = 10
spring.datasourceslave.druid.max-active = 20
spring.datasourceslave.druid.maxWait = 6000
spring.datasource.druid.validationQuery = SELECT 1 FROM DUAL

spring.datasource.druid.url = jdbc:mysql://127.0.0.1:3306/test?useUnicode=true&characterEncoding=utf8&characterSetResults=utf8&serverTimezone=GMT%2B8&useSSL=false
spring.datasource.druid.username = root
spring.datasource.druid.password = 123

分表

自定义分表规则，这里用创建时间截取年分表

UserComplexKeysShardingTableRule

package com.example.demo.config;

import java.text.SimpleDateFormat;
import java.util.*;

import org.apache.shardingsphere.api.sharding.complex.ComplexKeysShardingAlgorithm;
import org.apache.shardingsphere.api.sharding.complex.ComplexKeysShardingValue;
import org.springframework.util.CollectionUtils;

import com.google.common.collect.Range;

/**
 * 多列分片规则定义
 * @author admin
 * @version 1.0
 * @since 2023/01/05 15:15
 */
public class UserComplexKeysShardingTableRule implements ComplexKeysShardingAlgorithm<String> {

    private static final String TABLE_COLUMN_TIME = "create_time";

    private static final String TABLE_PREFIX      = "user_";

    @Override
    public Collection<String> doSharding(Collection<String> availableTargetNames, ComplexKeysShardingValue<String> shardingValue) {
        return getShardingValue(shardingValue);
    }

    /**
     * 获取分片键对应的值
     *
     * @param shardingValue shardingValue
     * @return Collection<String>
     */
    private Collection<String> getShardingValue(ComplexKeysShardingValue<String> shardingValue) {
        Collection<String> times = new HashSet<>();

        Map<String, Collection<String>> columnNameAndShardingValuesMap = shardingValue.getColumnNameAndShardingValuesMap();
        Map<String, Range<String>> columnNameAndRangeValuesMap = shardingValue.getColumnNameAndRangeValuesMap();

        if (columnNameAndShardingValuesMap.containsKey(TABLE_COLUMN_TIME)) {
            Collection<String> collection = columnNameAndShardingValuesMap.get(TABLE_COLUMN_TIME);
            Object next = collection.iterator().next();
            times.add(next.toString().substring(2, 4));
        }
        if (columnNameAndRangeValuesMap.containsKey(TABLE_COLUMN_TIME)) {
            Range<String> range = columnNameAndRangeValuesMap.get(TABLE_COLUMN_TIME);
            Collection<String> values = getTimeList(range);
            if (values.isEmpty()) {
                throw new UnsupportedOperationException("分片规则键不能为空");
            } else {
                times.addAll(values);
            }
        }
        if (CollectionUtils.isEmpty(times)) {
            throw new UnsupportedOperationException("分片规则键不能为空");
        }
        Set<String> tableNames = new HashSet<>();
        for (String billTime : times) {
            tableNames.add(getTableName(billTime));
        }
        return tableNames;
    }

    /**
     * 获取时间
     *
     * @param valueRange valueRange
     * @return Collection<String>
     */
    private Collection<String> getTimeList(Range<String> valueRange) {
        Set<String> prefixes = new HashSet<>();
        if (valueRange.isEmpty()) {
            throw new UnsupportedOperationException("分片规则键不能为空");
        } else {
            int start = Integer.parseInt(valueRange.lowerEndpoint().substring(2, 4));
            int end = Integer.parseInt(valueRange.upperEndpoint().substring(2, 4));
            int max = Integer.parseInt(dateToString(new Date(), "yy"));
            // 只能查2018年开始到当前年份的数据
            start = Math.max(start, 18);
            start = Math.min(start, max);
            end = Math.max(end, 18);
            end = Math.min(end, max);
            while (start <= end) {
                prefixes.add(String.valueOf(start));
                start++;
            }
            return prefixes;
        }
    }

    /**
     * 转格式
     * @param date date
     * @param format format
     * @return String
     */
    private String dateToString(Date date, String format) {
        SimpleDateFormat formater = new SimpleDateFormat(format);
        return formater.format(date);
    }

    /**
     * 根据学校编码分表
     *
     * @param simpleYear 年份的第3、4位
     * @return String
     */
    private static String getTableName(String simpleYear) {
        if (simpleYear.length() != 2) {
            throw new UnsupportedOperationException("分片规则键不能为空");
        }
        System.out.println("getTableName=" + TABLE_PREFIX + simpleYear);
        return TABLE_PREFIX + simpleYear;
    }

}

getColumnNameAndShardingValuesMap() // 分片键= shardingValue.getColumnNameAndRangeValuesMap();// 分片键IN和范围查询

验证分表

表是这样的

CREATE TABLE `user_21` (
  `id` int(10) NOT NULL,
  `name` varchar(32) DEFAULT NULL,
  `create_time` datetime DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4;

实体类和service这里就不赘述了，这里贴下查询TestController

    @RequestMapping("select")
    public String select() {
        LambdaQueryWrapper<User> queryWrapper = Wrappers.lambdaQuery();
        queryWrapper.eq(User::getCreateTime, new Date());
        userService.list(queryWrapper);
        return "ok";
    }

结果：日志打印

当前时间是二三年，这里的报错是因为没有创建对应表，创建user_23后则正常

常见问题

自定义分表规则未生效

场景: 分表规则字段范围查询且有拼接或格式处理

解决: 去掉sql上的字段格式化处理, 在服务层对传入参数进行处理

失效示例代码

and pay_time between DATE_FORMAT(#{param.billDate,jdbcType=VARCHAR}, '%Y-%m-%d 00:00:00')
                AND DATE_FORMAT(#{param.billDate,jdbcType=VARCHAR}, '%Y-%m-%d 23:59:59')

and pay_time between CONCAT(#{param.billDate,jdbcType=VARCHAR}, ' 00:00:00') AND
                CONCAT(#{param.billDate,jdbcType=VARCHAR}, ' 23:59:59')

AND DATE_FORMAT(pay_time, '%Y-%m-%d') = #{param.billDate,jdbcType=VARCHAR}

成功示例代码

and pay_time between #{param.payTimeStart,jdbcType=VARCHAR} AND #{param.payTimeEnd,jdbcType=VARCHAR}

持续更新ing!