温馨提示×

温馨提示×

您好,登录后才能下订单哦!

密码登录×
登录注册×
其他方式登录
点击 登录注册 即表示同意《亿速云用户服务条款》

sharding-jdbc中MasterSlaveRouter的作用是什么

发布时间:2021-06-22 15:12:07 来源:亿速云 阅读:310 作者:Leah 栏目:大数据

本篇文章为大家展示了sharding-jdbc中MasterSlaveRouter的作用是什么,内容简明扼要并且容易理解,绝对能使你眼前一亮,通过这篇文章的详细介绍希望你能有所收获。

MasterSlaveRouter

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-route/src/main/java/org/apache/shardingsphere/core/route/router/masterslave/MasterSlaveRouter.java

@RequiredArgsConstructor
public final class MasterSlaveRouter {
    
    private final MasterSlaveRule masterSlaveRule;
    
    private final boolean showSQL;
    
    /**
     * Route Master slave.
     *
     * @param sql SQL
     * @return data source names
     */
    // TODO for multiple masters may return more than one data source
    public Collection<String> route(final String sql) {
        Collection<String> result = route(new SQLJudgeEngine(sql).judge().getType());
        if (showSQL) {
            SQLLogger.logSQL(sql, result);
        }
        return result;
    }
    
    private Collection<String> route(final SQLType sqlType) {
        if (isMasterRoute(sqlType)) {
            MasterVisitedManager.setMasterVisited();
            return Collections.singletonList(masterSlaveRule.getMasterDataSourceName());
        }
        return Collections.singletonList(masterSlaveRule.getLoadBalanceAlgorithm().getDataSource(
                masterSlaveRule.getName(), masterSlaveRule.getMasterDataSourceName(), new ArrayList<>(masterSlaveRule.getSlaveDataSourceNames())));
    }
    
    private boolean isMasterRoute(final SQLType sqlType) {
        return SQLType.DQL != sqlType || MasterVisitedManager.isMasterVisited() || HintManager.isMasterRouteOnly();
    }
}
  • MasterSlaveRouter的route方法使用masterSlaveRule来进行路由

MasterSlaveRule

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-common/src/main/java/org/apache/shardingsphere/core/rule/MasterSlaveRule.java

@Getter
public class MasterSlaveRule {
    
    private final String name;
    
    private final String masterDataSourceName;
    
    private final Collection<String> slaveDataSourceNames;
    
    private final MasterSlaveLoadBalanceAlgorithm loadBalanceAlgorithm;
    
    private final MasterSlaveRuleConfiguration masterSlaveRuleConfiguration;
    
    public MasterSlaveRule(final String name, final String masterDataSourceName, final Collection<String> slaveDataSourceNames, final MasterSlaveLoadBalanceAlgorithm loadBalanceAlgorithm) {
        this.name = name;
        this.masterDataSourceName = masterDataSourceName;
        this.slaveDataSourceNames = slaveDataSourceNames;
        this.loadBalanceAlgorithm = null == loadBalanceAlgorithm ? new MasterSlaveLoadBalanceAlgorithmServiceLoader().newService() : loadBalanceAlgorithm;
        masterSlaveRuleConfiguration = new MasterSlaveRuleConfiguration(name, masterDataSourceName, slaveDataSourceNames, 
                new LoadBalanceStrategyConfiguration(this.loadBalanceAlgorithm.getType(), this.loadBalanceAlgorithm.getProperties()));
    }
    
    public MasterSlaveRule(final MasterSlaveRuleConfiguration config) {
        name = config.getName();
        masterDataSourceName = config.getMasterDataSourceName();
        slaveDataSourceNames = config.getSlaveDataSourceNames();
        loadBalanceAlgorithm = createMasterSlaveLoadBalanceAlgorithm(config.getLoadBalanceStrategyConfiguration());
        masterSlaveRuleConfiguration = config;
    }
    
    private MasterSlaveLoadBalanceAlgorithm createMasterSlaveLoadBalanceAlgorithm(final LoadBalanceStrategyConfiguration loadBalanceStrategyConfiguration) {
        MasterSlaveLoadBalanceAlgorithmServiceLoader serviceLoader = new MasterSlaveLoadBalanceAlgorithmServiceLoader();
        return null == loadBalanceStrategyConfiguration ? serviceLoader.newService() : serviceLoader.newService(loadBalanceStrategyConfiguration.getType(), loadBalanceStrategyConfiguration.getProperties());
    }
    
    /**
     * Judge whether contain data source name.
     *
     * @param dataSourceName data source name
     * @return contain or not.
     */
    public boolean containDataSourceName(final String dataSourceName) {
        return masterDataSourceName.equals(dataSourceName) || slaveDataSourceNames.contains(dataSourceName);
    }
}
  • MasterSlaveRule内置了loadBalanceAlgorithm、masterSlaveRuleConfiguration

MasterSlaveLoadBalanceAlgorithm

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-api/src/main/java/org/apache/shardingsphere/spi/masterslave/MasterSlaveLoadBalanceAlgorithm.java

public interface MasterSlaveLoadBalanceAlgorithm extends TypeBasedSPI {
    
    /**
     * Get data source.
     * 
     * @param name master-slave logic data source name
     * @param masterDataSourceName name of master data sources
     * @param slaveDataSourceNames names of slave data sources
     * @return name of selected data source
     */
    String getDataSource(String name, String masterDataSourceName, List<String> slaveDataSourceNames);
}
  • MasterSlaveLoadBalanceAlgorithm接口继承了TypeBasedSPI接口,它定义了getDataSource方法;它有两个实现类分别是RandomMasterSlaveLoadBalanceAlgorithm、RoundRobinMasterSlaveLoadBalanceAlgorithm

RandomMasterSlaveLoadBalanceAlgorithm

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-common/src/main/java/org/apache/shardingsphere/core/strategy/masterslave/RandomMasterSlaveLoadBalanceAlgorithm.java

@Getter
@Setter
public final class RandomMasterSlaveLoadBalanceAlgorithm implements MasterSlaveLoadBalanceAlgorithm {
    
    private Properties properties = new Properties();
    
    @Override
    public String getType() {
        return "RANDOM";
    }
    
    @Override
    public String getDataSource(final String name, final String masterDataSourceName, final List<String> slaveDataSourceNames) {
        return slaveDataSourceNames.get(new Random().nextInt(slaveDataSourceNames.size()));
    }
}
  • RandomMasterSlaveLoadBalanceAlgorithm使用Random().nextInt来进行随机

RoundRobinMasterSlaveLoadBalanceAlgorithm

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-common/src/main/java/org/apache/shardingsphere/core/strategy/masterslave/RoundRobinMasterSlaveLoadBalanceAlgorithm.java

@Getter
@Setter
public final class RoundRobinMasterSlaveLoadBalanceAlgorithm implements MasterSlaveLoadBalanceAlgorithm {
    
    private static final ConcurrentHashMap<String, AtomicInteger> COUNTS = new ConcurrentHashMap<>();
    
    private Properties properties = new Properties();
    
    @Override
    public String getType() {
        return "ROUND_ROBIN";
    }
    
    @Override
    public String getDataSource(final String name, final String masterDataSourceName, final List<String> slaveDataSourceNames) {
        AtomicInteger count = COUNTS.containsKey(name) ? COUNTS.get(name) : new AtomicInteger(0);
        COUNTS.putIfAbsent(name, count);
        count.compareAndSet(slaveDataSourceNames.size(), 0);
        return slaveDataSourceNames.get(Math.abs(count.getAndIncrement()) % slaveDataSourceNames.size());
    }
}
  • RoundRobinMasterSlaveLoadBalanceAlgorithm使用Math.abs(count.getAndIncrement()) % slaveDataSourceNames.size()进行round robin

MasterSlaveRuleConfiguration

incubator-shardingsphere-4.0.0-RC1/sharding-core/sharding-core-api/src/main/java/org/apache/shardingsphere/api/config/masterslave/MasterSlaveRuleConfiguration.java

@Getter
public class MasterSlaveRuleConfiguration implements RuleConfiguration {
    
    private final String name;
    
    private final String masterDataSourceName;
    
    private final Collection<String> slaveDataSourceNames;
    
    private final LoadBalanceStrategyConfiguration loadBalanceStrategyConfiguration;
    
    public MasterSlaveRuleConfiguration(final String name, final String masterDataSourceName, final Collection<String> slaveDataSourceNames) {
        this(name, masterDataSourceName, slaveDataSourceNames, null);
    }
    
    public MasterSlaveRuleConfiguration(final String name, 
                                        final String masterDataSourceName, final Collection<String> slaveDataSourceNames, final LoadBalanceStrategyConfiguration loadBalanceStrategyConfiguration) {
        Preconditions.checkArgument(!Strings.isNullOrEmpty(name), "Name is required.");
        Preconditions.checkArgument(!Strings.isNullOrEmpty(masterDataSourceName), "MasterDataSourceName is required.");
        Preconditions.checkArgument(null != slaveDataSourceNames && !slaveDataSourceNames.isEmpty(), "SlaveDataSourceNames is required.");
        this.name = name;
        this.masterDataSourceName = masterDataSourceName;
        this.slaveDataSourceNames = slaveDataSourceNames;
        this.loadBalanceStrategyConfiguration = loadBalanceStrategyConfiguration;
    }
}
  • MasterSlaveRuleConfiguration定义了name、masterDataSourceName、slaveDataSourceNames、loadBalanceStrategyConfiguration属性

小结

MasterSlaveRouter的route方法使用masterSlaveRule来进行路由;MasterSlaveRule内置了loadBalanceAlgorithm、masterSlaveRuleConfiguration;MasterSlaveLoadBalanceAlgorithm接口继承了TypeBasedSPI接口,它定义了getDataSource方法;它有两个实现类分别是RandomMasterSlaveLoadBalanceAlgorithm、RoundRobinMasterSlaveLoadBalanceAlgorithm

上述内容就是sharding-jdbc中MasterSlaveRouter的作用是什么,你们学到知识或技能了吗?如果还想学到更多技能或者丰富自己的知识储备,欢迎关注亿速云行业资讯频道。

向AI问一下细节

免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。

AI