使用雪花算法生成分布式唯一ID

发布于:2025-04-12 ⋅ 阅读:(55) ⋅ 点赞:(0)

雪花算法(Snowflake)原理详解

雪花算法是由Twitter提出的一种分布式唯一ID生成算法,旨在为每个请求生成一个全局唯一的64位整数ID。这个ID由以下几个部分组成:

  1. 符号位:1位,始终为0,保证生成的ID为正数。
  2. 时间戳:41位,表示自定义纪元以来的时间戳(毫秒级别),可以提供约69年的时间跨度。
  3. 数据中心ID:5位,支持最多31个数据中心。
  4. 机器ID:5位,在同一数据中心内区分不同的机器或服务实例,支持最多31个实例。
  5. 序列号:12位,用于在同一毫秒内生成的多个ID之间进行区分,最大值为4095。

因此,一个典型的Snowflake ID结构如下:

0 - 41 bits for timestamp - 5 bits for data center id - 5 bits for worker id - 12 bits for sequence number
时间回拨处理

由于Snowflake依赖于系统时间来生成ID,因此如果服务器时间被调整(例如向后调整),可能会导致生成重复的ID。为了防止这种情况,通常会采取以下措施之一:

  • 等待直到系统时间超过上次记录的时间戳。
  • 使用序列号临时解决轻微的时间回拨问题。

在Spring Boot和MyBatis中的实现与集成

SnowflakeIdWorker.java
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

public class SnowflakeIdWorker {
    private static final Logger logger = LoggerFactory.getLogger(SnowflakeIdWorker.class);

    // 自定义纪元时间戳,这里设置为当前时间
    private final long twepoch = 1712636400000L; // 2025-04-09 00:00:00 UTC 时间戳

    // 机器id所占的位数
    private final long workerIdBits = 5L;
    // 数据标识id所占的位数
    private final long datacenterIdBits = 5L;
    // 支持的最大机器id,结果是31 (这个值的结果应该是31,因为max值是1左移5位再减1,也就是(2^5)-1=31)
    private final long maxWorkerId = -1L ^ (-1L << workerIdBits);
    // 支持的最大数据标识id,结果也是31
    private final long maxDatacenterId = -1L ^ (-1L << datacenterIdBits);
    // 序列在id中占的位数
    private final long sequenceBits = 12L;

    // 工作机器ID(0~31)
    private final long workerId;
    // 数据中心ID(0~31)
    private final long datacenterId;
    // 毫秒内序列(0~4095)
    private volatile long sequence = 0L;
    // 上次生成ID的时间截
    private volatile long lastTimestamp = -1L;

    /**
     * 构造函数,初始化workerId和datacenterId
     * @param workerId 工作机器ID
     * @param datacenterId 数据中心ID
     */
    public SnowflakeIdWorker(long workerId, long datacenterId) {
        if (workerId > maxWorkerId || workerId < 0) {
            throw new IllegalArgumentException(String.format("worker Id can't be greater than %d or less than 0", maxWorkerId));
        }
        if (datacenterId > maxDatacenterId || datacenterId < 0) {
            throw new IllegalArgumentException(String.format("datacenter Id can't be greater than %d or less than 0", maxDatacenterId));
        }
        this.workerId = workerId;
        this.datacenterId = datacenterId;
    }

    /**
     * 获取下一个ID
     * @return 下一个全局唯一的ID
     */
    public synchronized long nextId() {
        long timestamp = timeGen();

        // 如果当前时间小于上次ID生成的时间戳,说明系统时钟回退过这个时候应当抛出异常
        if (timestamp < lastTimestamp) {
            logger.error(String.format("Clock moved backwards. Refusing to generate id for %d milliseconds", lastTimestamp - timestamp));
            throw new RuntimeException(String.format("Clock moved backwards. Refusing to generate id for %d milliseconds", lastTimestamp - timestamp));
        }

        // 如果是同一时间生成的,则进行毫秒内序列
        if (lastTimestamp == timestamp) {
            sequence = (sequence + 1) & ((1L << sequenceBits) - 1);
            // 毫秒内序列溢出
            if (sequence == 0) {
                // 阻塞到下一个毫秒,获得新的时间戳
                timestamp = tilNextMillis(lastTimestamp);
            }
        } else {
            // 时间戳改变,毫秒内序列重置
            sequence = 0L;
        }

        // 上次生成ID的时间截
        lastTimestamp = timestamp;

        // 移位并通过或运算拼到一起组成64位的ID
        return ((timestamp - twepoch) << (sequenceBits + workerIdBits + datacenterIdBits)) |
               (datacenterId << (sequenceBits + workerIdBits)) |
               (workerId << sequenceBits) |
               sequence;
    }

    /**
     * 阻塞到下一个毫秒,直到获得新的时间戳
     * @param lastTimestamp 上次生成ID的时间戳
     * @return 当前时间戳
     */
    protected long tilNextMillis(long lastTimestamp) {
        long timestamp = timeGen();
        while (timestamp <= lastTimestamp) {
            timestamp = timeGen();
        }
        return timestamp;
    }

    /**
     * 返回以毫秒为单位的当前时间
     * @return 当前时间(毫秒)
     */
    protected long timeGen() {
        return System.currentTimeMillis();
    }
}
SnowflakeConfig.java
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

@Configuration
public class SnowflakeConfig {

    /**
     * 创建SnowflakeIdWorker Bean
     * @return SnowflakeIdWorker实例
     */
    @Bean
    public SnowflakeIdWorker snowflakeIdWorker() {
        // 根据实际情况调整workerId和datacenterId
        return new SnowflakeIdWorker(1, 1); // 这里根据你的实际环境配置
    }
}
UserMapper.java
import org.apache.ibatis.annotations.Insert;
import org.apache.ibatis.annotations.Mapper;
import org.apache.ibatis.annotations.Param;

@Mapper
public interface UserMapper {
    /**
     * 插入用户信息
     * @param id 用户ID
     * @param name 用户名
     * @param email 用户邮箱
     */
    @Insert("INSERT INTO users(id, name, email) VALUES(#{id}, #{name}, #{email})")
    void insertUser(@Param("id") Long id, @Param("name") String name, @Param("email") String email);
}
UserService.java
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Service;

@Service
public class UserService {

    @Autowired
    private UserMapper userMapper;

    @Autowired
    private SnowflakeIdWorker snowflakeIdWorker;

    /**
     * 创建用户
     * @param name 用户名
     * @param email 用户邮箱
     */
    public void createUser(String name, String email) {
        // 使用Snowflake算法生成唯一ID
        Long id = snowflakeIdWorker.nextId();
        // 调用Mapper插入用户信息
        userMapper.insertUser(id, name, email);
    }
}

注意事项与最佳实践

  1. 时间同步

    • 确保所有服务器的时间同步非常重要。建议使用NTP服务来保证所有节点的时间一致性。
  2. workerId和datacenterId的分配

    • 必须确保每个服务实例的workerIddatacenterId是唯一的,以避免ID冲突。
  3. 性能优化

    • 对于高并发场景,考虑减少锁的粒度或者采用无锁编程技术。
    • 可以考虑批量ID生成策略,即预先生成一批ID并在内存中缓存起来供后续请求使用,从而减少对全局锁的竞争。
  4. 日志记录与监控

    • 增加详细的日志记录,便于调试和维护。
    • 设置监控指标,如生成ID的速度、失败率等,并在出现问题时触发报警。可以使用Prometheus、Grafana等工具进行监控。

网站公告

今日签到

点亮在社区的每一天
去签到