接口限流实现方式及算法

实现方案

1.基于guava 限流实现（单机版）

guava 为谷歌开源的一个比较实用的组件，利用这个组件可以帮助开发人员完成常规的限流操作，接下来看具体的实现步骤

1.1 依赖引入

<dependency>
    <groupId>com.google.guava</groupId>
    <artifactId>guava</artifactId>
    <version>30.1-jre</version>
</dependency>

1.2 自定义注解

package com.zhen.studytotal.limitRequest.annotation;

import java.lang.annotation.*;

/**
 * Description: 自定义限流注解-基于gvava
 * User: zhen0w0
 * Date: 2025/3/14
 * Time: 00:32
 */
@Documented
@Target(value = ElementType.METHOD)
@Retention(value = RetentionPolicy.RUNTIME)
public @interface GuavaLimitRateAnnotation {

    //限制类型
    String limitType();

    //每秒五个请求
    double limitCount() default 5d;
}

1.3 guava工具类

package com.zhen.studytotal.limitRequest.aspect;
 
import com.google.common.util.concurrent.RateLimiter;

import java.util.HashMap;
import java.util.Map;

/**
 * @description 创建RateLimiter
 * @author tcm
 * @version 1.0.0
 * @date 2021/10/12 17:55
 **/
public class RateLimiterHelper {

    private RateLimiterHelper(){}

    //根据mapkey，对不同注解类型限流
    private static Map<String, RateLimiter> rateMap = new HashMap<>();

    public static RateLimiter getRateLimiter(String limitType, double limitCount ){
        RateLimiter rateLimiter = rateMap.get(limitType);
        if(rateLimiter == null){
            rateLimiter = RateLimiter.create(limitCount);
            rateMap.put(limitType,rateLimiter);
        }
        return rateLimiter;
    }
 
}

1.4 AOP切面类

package com.zhen.studytotal.limitRequest.aspect;


import com.alibaba.fastjson2.JSONObject;
import com.fasterxml.jackson.databind.util.JSONPObject;
import com.google.common.util.concurrent.RateLimiter;
import com.zhen.studytotal.limitRequest.annotation.GuavaLimitRateAnnotation;
import org.aspectj.lang.JoinPoint;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Before;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.stereotype.Component;
import org.springframework.web.context.request.RequestContextHolder;
import org.springframework.web.context.request.ServletRequestAttributes;


import javax.servlet.ServletOutputStream;
import javax.servlet.http.HttpServletResponse;
import java.io.IOException;
import java.lang.reflect.Method;
import java.util.Objects;

/**
 * Description: 接口限流切面类
 * User: zhen0w0
 * Date: 2025/3/15
 * Time: 02:20
 */
@Aspect
@Component
public class GuavaLimitRateAspect {

    private static Logger logger = LoggerFactory.getLogger(GuavaLimitRateAspect.class);

    @Before("execution(@com.zhen.studytotal.limitRequest.annotation.GuavaLimitRateAnnotation * *(..))")
    public void limit(JoinPoint joinPoint){
        //1.获取当前方法
        Method currentMethod = getCurrentMethod(joinPoint);
        if(Objects.isNull(currentMethod)){
            return;
        }
        //2.获取注解参数
        String limitType = currentMethod.getAnnotation(GuavaLimitRateAnnotation.class).limitType();
        double limitCount = currentMethod.getAnnotation(GuavaLimitRateAnnotation.class).limitCount();
        //3.使用guava令牌桶算法获取令牌，获取不到先等待
        RateLimiter rateLimiter = RateLimiterHelper.getRateLimiter(limitType, limitCount);
        boolean b = rateLimiter.tryAcquire();
        if (b){
            logger.info("获取令牌成功");
        } else {
            HttpServletResponse response = ((ServletRequestAttributes) RequestContextHolder.getRequestAttributes()).getResponse();
            //这里使用的json第三方依赖为（fastjson2）
            JSONObject jsonObject = new JSONObject();
            jsonObject.put("success", false);
            jsonObject.put("message", "请求频繁，限流中···");
            try {
                output(response, jsonObject.toJSONString());
            }catch (Exception e){
                logger.error("限流异常:{}", e);
            }

        }

    }

    private void output(HttpServletResponse response, String jsonString) throws IOException {
        response.setContentType("application/json;charset=UTF-8");
        ServletOutputStream outputStream = null;
        try {
            outputStream = response.getOutputStream();
            outputStream.write(jsonString.getBytes("UTF-8"));
        }catch (IOException e){
            e.printStackTrace();
        }finally {
            assert outputStream != null;
            outputStream.flush();
            outputStream.close();
        }
    }

    private Method getCurrentMethod(JoinPoint joinPoint) {
        Method[] methods = joinPoint.getTarget().getClass().getMethods();
        Method target = null;
        for (Method method : methods) {
            if (method.getName().equals(joinPoint.getSignature().getName())) {
                target = method;
                break;
            }
        }
        return target;
    }


}

1.5 测试

package com.zhen.studytotal.limitRequest.controller;

import com.zhen.studytotal.limitRequest.annotation.GuavaLimitRateAnnotation;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RestController;

/**
 * Description: 限流测试
 * User: zhen0w0
 * Date: 2025/3/16
 * Time: 02:54
 */
@RestController
@RequestMapping("/test/guava")
public class TestGuavaController {

    @GetMapping("/limit")
    @GuavaLimitRateAnnotation(limitType = "测试guava限流", limitCount = 1)
    public String test(){
        return "limit";
    }
}

1.6 效果

这里为了清楚看到效果，限流QPS设置为1，多刷新几次请求即可

2. 基于 redis+lua 限流实现（分布式版）

redis是线程安全的，天然具有线程安全的特性，支持原子性操作，限流服务不仅需要承接超高QPS，还要保证限流逻辑的执行层面具备线程安全的特性，利用Redis这些特性做限流，既能保证线程安全，也能保证性能

结合流程图可以得出以下实现思路：

1.编写 lua 脚本，指定入参的限流规则，比如对特定的接口限流时，可以根据某个或几个参数进行判定，调用该接口的请求，在一定的时间窗口内监控请求次数；
2.既然是限流，最好能够通用，可将限流规则应用到任何接口上，那么最合适的方式就是通过自定义注解形式切入；
3.提供一个配置类，被 spring 的容器管理，redisTemplate 中提供了 DefaultRedisScript这个 bean；
4.提供一个能动态解析接口参数的类，根据接口参数进行规则匹配后触发限流；

2.1 依赖

YML文件记得添加redis相关配置

 <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-data-redis</artifactId>
</dependency>

2.2 自定义注解

package com.zhen.studytotal.limitRequest.annotation;

import com.zhen.studytotal.limitRequest.enums.LimitTypeEnum;
import org.apache.commons.lang3.StringUtils;

import java.lang.annotation.*;

/**
  * Description: redis限流注解
  * User: zhen0w0
  * Date: 2025/3/17
  * Time: 00:11
  */
@Documented
@Target(value = ElementType.METHOD)
@Retention(value = RetentionPolicy.RUNTIME)
public @interface RedisLimitAnnotation {

    /**
     * redis-key
     */
    String key() default "";

    /**
     * redis-key前缀
     */
    String prefix() default "";

    /**
     * 限流时间内限流次数
     */
    int count();

    /**
     * 限流时间-单位秒
     */
    int time();

    /**
     * 限流类型
     */
    LimitTypeEnum limitType() default LimitTypeEnum.INTERFACE;
}

2.3 限流类型枚举

package com.zhen.studytotal.limitRequest.enums;

import lombok.Getter;

/**
 * Description:
 * User: zhen0w0
 * Date: 2025/3/17
 * Time: 00:35
 */
@Getter
public enum LimitTypeEnum {

    // 默认限流策略，针对某一个接口进行限流
    INTERFACE
    ,

    // 根据IP地址进行限流
    IP
    ,

    // 自定义的Key
    CUSTOMER
    ;
}

2.4 IP工具类

package com.zhen.studytotal.limitRequest.utils;

import javax.servlet.http.HttpServletRequest;
import java.net.HttpCookie;
import java.net.InetAddress;
import java.net.UnknownHostException;

/**
 * Description: iputils
 * User: zhen0w0
 * Date: 2025/3/17
 * Time: 23:08
 */
public class IPUtils {

    public static String getIpAddr(HttpServletRequest request) {
        String ip = request.getHeader("x-forwarded-for");
        if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
            ip = request.getHeader("Proxy-Client-IP");
        }
        if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
            ip = request.getHeader("WL-Proxy-Client-IP");
        }
        if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
            ip = request.getHeader("HTTP_CLIENT_IP");
        }
        if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
            ip = request.getHeader("HTTP_X_FORWARDED_FOR");
        }
        if (ip == null || ip.length() == 0 || "unknown".equalsIgnoreCase(ip)) {
            ip = request.getRemoteAddr();
        }

        // 本机访问
        if ("localhost".equalsIgnoreCase(ip) || "127.0.0.1".equalsIgnoreCase(ip) || "0:0:0:0:0:0:0:1".equalsIgnoreCase(ip)){
            // 根据网卡取本机配置的IP
            InetAddress inet;
            try {
                inet = InetAddress.getLocalHost();
                ip = inet.getHostAddress();
            } catch (UnknownHostException e) {
                e.printStackTrace();
            }
        }
        // 对于通过多个代理的情况，第一个IP为客户端真实IP,多个IP按照','分割
        //"***.***.***.***".length() = 15
        if (null != ip && ip.length() > 15) {
            if (ip.indexOf(",") > 15) {
                ip = ip.substring(0, ip.indexOf(","));
            }
        }
        return ip;
    }

}

2.5 自定义 lua 脚本

-- 定义变量：redis中key值、规定的时间段内访问次数、redis中过期时间、当前访问次数

local key = KEYS[1]
local limit = tonumber(ARGV[1])
local count = tonumber(ARGV[2])
local current = tonumber(redis.call('get', key) or "0")

if current + 1 > limit then
  return 0
end
   -- 没有超阈值，将当前访问数量+1，
   current = redis.call("INCRBY", key, "1")
if tonumber(current) == 1 then
   -- 设置过期时间
   redis.call("expire", key, count)
end
   return current

2.6 Redis配置类

设置执行lua脚本，这里注意lua脚本返回类型，我这里原来采用number类型接收返回，报错

io.lettuce.core.output.ValueOutput does not support set(long)，改用Long类型

package com.zhen.studytotal.limitRequest.config;

import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.core.io.ClassPathResource;
import org.springframework.data.redis.connection.RedisConnectionFactory;
import org.springframework.data.redis.core.RedisTemplate;
import org.springframework.data.redis.core.script.DefaultRedisScript;
import org.springframework.data.redis.serializer.Jackson2JsonRedisSerializer;
import org.springframework.data.redis.serializer.StringRedisSerializer;
import org.springframework.scripting.support.ResourceScriptSource;

/**
 * Description: redis配置
 * User: zhen0w0
 * Date: 2025/3/17
 * Time: 00:47
 */
@Configuration
public class RedisConfiguration {

    @Bean
    public DefaultRedisScript<Long> redisLuaScript(){
        DefaultRedisScript<Long> numberDefaultRedisScript = new DefaultRedisScript<>();
        numberDefaultRedisScript.setScriptSource(new ResourceScriptSource(new ClassPathResource("lua\\limit.lua")));
        //设置lua脚本返回值类型，需要和lua脚本的返回值类型一致
        numberDefaultRedisScript.setResultType(Long.class);
        return numberDefaultRedisScript;
    }

    @Bean("redisTemplate")
    public RedisTemplate<String, Object> redisTemplate(RedisConnectionFactory redisConnectionFactory) {
        RedisTemplate<String, Object> redisTemplate= new RedisTemplate<>();
        redisTemplate.setConnectionFactory(redisConnectionFactory);
        Jackson2JsonRedisSerializer jackson2JsonRedisSerializer = new Jackson2JsonRedisSerializer(Object.class);
        //设置value的序列化方式为json
        redisTemplate.setValueSerializer(jackson2JsonRedisSerializer);
        //设置key的序列化方式为string
        redisTemplate.setKeySerializer(new StringRedisSerializer());
        redisTemplate.setHashKeySerializer(new StringRedisSerializer());
        redisTemplate.setHashValueSerializer(jackson2JsonRedisSerializer);
        redisTemplate.afterPropertiesSet();
        return redisTemplate;
    }
}

2.7 Redis切面类

package com.zhen.studytotal.limitRequest.aspect;

import com.zhen.studytotal.limitRequest.annotation.RedisLimitAnnotation;
import com.zhen.studytotal.limitRequest.enums.LimitTypeEnum;
import com.zhen.studytotal.limitRequest.utils.IPUtils;
import org.apache.commons.lang3.StringUtils;
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.Signature;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Pointcut;
import org.aspectj.lang.reflect.MethodSignature;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.data.redis.core.RedisTemplate;
import org.springframework.data.redis.core.script.DefaultRedisScript;
import org.springframework.stereotype.Component;
import org.springframework.web.context.request.RequestContextHolder;
import org.springframework.web.context.request.ServletRequestAttributes;

import javax.annotation.Resource;
import javax.servlet.http.HttpServletRequest;
import java.lang.reflect.Method;
import java.util.Collections;
import java.util.Objects;

/**
 * Description: reids限流切面
 * User: zhen0w0
 * Date: 2025/3/17
 * Time: 01:05
 */
@Aspect
@Component
public class RedisLimitAspect {
    public static final Logger logger = LoggerFactory.getLogger(RedisLimitAspect.class);

    @Autowired
    private RedisTemplate<String, Object> redisTemplate;

    @Autowired
    private DefaultRedisScript<Long> redisLuaScript;

    @Pointcut("@annotation(com.zhen.studytotal.limitRequest.annotation.RedisLimitAnnotation)")
    public void redisLimit(){}

    @Around("redisLimit()")
    public Object interceptor(ProceedingJoinPoint joinPoint) throws Throwable {
        MethodSignature signature = (MethodSignature) joinPoint.getSignature();
        Method method = signature.getMethod();
        RedisLimitAnnotation annotation = method.getAnnotation(RedisLimitAnnotation.class);
        if (Objects.isNull(annotation)){
            return joinPoint.proceed();
        }

        String key = getKeyByLimitType(annotation,signature);
        //调用lua脚本，获取返回结果，这里即为请求次数
        Long number = redisTemplate.execute(redisLuaScript, Collections.singletonList(key), annotation.count(), annotation.time());
        logger.info("限流时时间内访问次数：{}",number);
        if (number != null  && number.intValue() != 0 && number.intValue() <= annotation.count()){
            logger.info("限流时时间内访问次数：{}",number);
            return joinPoint.proceed();
        }
        throw new RuntimeException("访问次数超过限制,限流中···");
    }

    /**
     *  三种redis-key
     *  1.自定义key：prefix + : + key
     *  2.接口限流：prefix + : + 接口全类名
     *  3.根据IP限流：prefix + : + IP地址 + - + 接口全类名
     * @param annotation
     * @param signature
     * @return
     */
    private String getKeyByLimitType(RedisLimitAnnotation annotation, MethodSignature signature) {
        String key = "";
        LimitTypeEnum limitTypeEnum = annotation.limitType();
        String prefix = annotation.prefix();
        if (StringUtils.isNotBlank(prefix)){
            key += prefix + ":";
        }

        if (LimitTypeEnum.CUSTOMER == limitTypeEnum) {
            String tempKey = annotation.key();
            if (StringUtils.isBlank(tempKey)) {
                throw new RuntimeException("redis限流->自定义类型，key不能为空");
            }
            return key + tempKey;
        }

        Method method = signature.getMethod();
        Class<?> targetClass = method.getDeclaringClass();
        String classFullName = targetClass.getName() + "-" + method.getName();
        if (LimitTypeEnum.INTERFACE == limitTypeEnum) {
            return key + classFullName;
        }
        //IP
        HttpServletRequest request = ((ServletRequestAttributes) RequestContextHolder.getRequestAttributes()).getRequest();
        String ipAddr = IPUtils.getIpAddr(request);
        return key + ipAddr + "-" + classFullName;
    }


}

2.8 测试

@GetMapping("/redis")
@RedisLimitAnnotation(key = "limitByRedis",time = 5,count = 5,limitType = LimitTypeEnum.IP)
public String testRedis(){
    return "REDIS";
}

3.基于 sentinel 限流实现（分布式版）

sentinel 通常是需要结合 springcloud-alibaba 框架一起实用的，而且与框架集成之后，可以配合控制台一起使用达到更好的效果，实际上，sentinel 官方也提供了相对原生的 SDK 可供使用，接下来就以这种方式进行整合

3.1 依赖

<dependency>
    <groupId>com.alibaba.csp</groupId>
    <artifactId>sentinel-core</artifactId>
    <version>1.8.0</version>
</dependency>

3.2 自定义注解

package com.zhen.studytotal.limitRequest.annotation;

import java.lang.annotation.*;

/**
 * Description:
 * User: zhen0w0
 * Date: 2025/3/19
 * Time: 00:11
 */
@Documented
@Target(value = ElementType.METHOD)
@Retention(value = RetentionPolicy.RUNTIME)
public @interface SentinelLimitRateAnnotation {

    //限制类型
    String resourceName();

    //每秒允许请求数量
    int limitCount() default 5;
}

3.3 AOP切面

package com.zhen.studytotal.limitRequest.aspect;

import com.alibaba.csp.sentinel.Entry;
import com.alibaba.csp.sentinel.SphU;
import com.alibaba.csp.sentinel.Tracer;
import com.alibaba.csp.sentinel.slots.block.BlockException;
import com.alibaba.csp.sentinel.slots.block.RuleConstant;
import com.alibaba.csp.sentinel.slots.block.flow.FlowRule;
import com.alibaba.csp.sentinel.slots.block.flow.FlowRuleManager;
import com.zhen.studytotal.limitRequest.annotation.SentinelLimitRateAnnotation;
import org.apache.commons.lang3.StringUtils;
import org.aspectj.lang.JoinPoint;
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Pointcut;
import org.checkerframework.checker.units.qual.A;
import org.springframework.stereotype.Component;

import java.lang.reflect.Method;
import java.util.ArrayList;
import java.util.List;
import java.util.Objects;

/**
 * Description:
 * User: zhen0w0
 * Date: 2025/3/19
 * Time: 00:20
 */
@Aspect
@Component
public class SentinelLimitRateAspect {

    @Pointcut(value = "@annotation(com.zhen.studytotal.limitRequest.annotation.SentinelLimitRateAnnotation)")
    public void rateLimit(){

    }

    @Around(value = "rateLimit()")
    public Object around(ProceedingJoinPoint joinPoint){
        // 1.获取当前方法
        Method currentMethod = getCurrentMethod(joinPoint);
        if (Objects.isNull(currentMethod)) {
            return null;
        }
        // 2.从方法注解定义上获取限流的类型
        String resourceName = currentMethod.getAnnotation(SentinelLimitRateAnnotation.class).resourceName();
        if(StringUtils.isEmpty(resourceName)){
            throw new RuntimeException("资源名称为空");
        }
        int limitCount = currentMethod.getAnnotation(SentinelLimitRateAnnotation.class).limitCount();
        // 3.初始化规则
        initFlowRule(resourceName,limitCount);
        Entry entry = null;
        Object result = null;
        try {
            entry = SphU.entry(resourceName);
            try {
                result = joinPoint.proceed();
            } catch (Throwable throwable) {
                throwable.printStackTrace();
            }
        } catch (BlockException ex) {
            // 资源访问阻止，被限流或被降级，在此处进行相应的处理操作
            System.out.println("blocked");
            return "被限流了";
        } catch (Exception e) {
            Tracer.traceEntry(e, entry);
        } finally {
            if (entry != null) {
                entry.exit();
            }
        }
        return result;
    }

    private static void initFlowRule(String resourceName,int limitCount) {
        List<FlowRule> rules = new ArrayList<>();
        FlowRule rule = new FlowRule();
        //设置受保护的资源
        rule.setResource(resourceName);
        //设置流控规则 QPS
        rule.setGrade(RuleConstant.FLOW_GRADE_QPS);
        //设置受保护的资源阈值
        rule.setCount(limitCount);
        rules.add(rule);
        //加载配置好的规则
        FlowRuleManager.loadRules(rules);
    }

    private Method getCurrentMethod(JoinPoint joinPoint) {
        Method[] methods = joinPoint.getTarget().getClass().getMethods();
        Method target = null;
        for (Method method : methods) {
            if (method.getName().equals(joinPoint.getSignature().getName())) {
                target = method;
                break;
            }
        }
        return target;
    }
}

3.4 测试

@GetMapping("/sentinel")
@SentinelLimitRateAnnotation(resourceName = "测试sentinel限流", limitCount = 1)
public String testSentinel(){
    return "Sentinel";
}

4.参考

https://blog.csdn.net/truelove12358/article/details/127751003