快捷導(dǎo)航

Java實(shí)現(xiàn)限流的6種方案詳解

更新時(shí)間：2025年06月08日 09:17:11 作者：VipSoft

在 Java 生態(tài)中,除了 Guava 的 RateLimiter,還有多種限流方案可供選擇,本文為大家整理了6個(gè)常用的方案,有需要的小伙伴可以根據(jù)需求進(jìn)行選擇

1. Spring Cloud Gateway / Spring Cloud Alibaba Sentinel

適用于： Spring Cloud 微服務(wù)架構(gòu)

// 在Spring Cloud Gateway中的配置
@Bean
public RouteLocator customRouteLocator(RouteLocatorBuilder builder) {
    return builder.routes()
        .route("qrcode_route", r -> r.path("/api/qrcode/**")
            .filters(f -> f.requestRateLimiter()
                .rateLimiter(RedisRateLimiter.class, config -> {
                    config.setBurstCapacity(20);
                    config.setReplenishRate(10);
                }))
            .uri("http://localhost:8080"))
        .build();
}

2. Resilience4j RateLimiter

適用于： 需要更豐富熔斷限流功能的場(chǎng)景

// 添加依賴
implementation 'io.github.resilience4j:resilience4j-ratelimiter:1.7.1'

// 使用示例
RateLimiterConfig config = RateLimiterConfig.custom()
    .limitRefreshPeriod(Duration.ofSeconds(1))
    .limitForPeriod(10)
    .timeoutDuration(Duration.ofMillis(100))
    .build();

RateLimiter rateLimiter = RateLimiter.of("qrcodeLimiter", config);

CheckedRunnable restrictedCall = RateLimiter
    .decorateCheckedRunnable(rateLimiter, () -> generateQrcode());

Try.run(restrictedCall)
    .onFailure(throwable -> response.sendError(429, "請(qǐng)求過(guò)于頻繁"));

3. Bucket4j

適用于： 需要分布式限流的場(chǎng)景

// 添加依賴
implementation 'com.github.vladimir-bukhtoyarov:bucket4j-core:7.0.0'

// 本地限流示例
Bandwidth limit = Bandwidth.classic(10, Refill.intervally(10, Duration.ofSeconds(1)));
Bucket bucket = Bucket.builder().addLimit(limit).build();

if (bucket.tryConsume(1)) {
    // 處理請(qǐng)求
} else {
    response.sendError(429, "請(qǐng)求過(guò)于頻繁");
}

4. Redis + Lua 分布式限流

適用于： 分布式環(huán)境下的精確限流

// Redis限流腳本
private static final String LIMIT_SCRIPT =
    "local key = KEYS[1]\n" +
    "local limit = tonumber(ARGV[1])\n" +
    "local expire = tonumber(ARGV[2])\n" +
    "local current = tonumber(redis.call('get', key) or 0\n" +
    "if current + 1 > limit then\n" +
    "    return 0\n" +
    "else\n" +
    "    redis.call('INCR', key)\n" +
    "    if current == 0 then\n" +
    "        redis.call('EXPIRE', key, expire)\n" +
    "    end\n" +
    "    return 1\n" +
    "end";

public boolean tryAcquire(String key, int limit, int expireSec) {
    Long result = redisTemplate.execute(
        new DefaultRedisScript<>(LIMIT_SCRIPT, Long.class),
        Collections.singletonList(key),
        String.valueOf(limit), String.valueOf(expireSec));
    return result != null && result == 1;
}

// 使用
if (!redisLimiter.tryAcquire("qrcode:"+ip, 10, 60)) {
    response.sendError(429, "請(qǐng)求過(guò)于頻繁");
    return;
}

5. Spring AOP 實(shí)現(xiàn)簡(jiǎn)單限流

適用于： 簡(jiǎn)單的單體應(yīng)用限流

@Aspect
@Component
public class RateLimitAspect {
    
    private final Map<String, AtomicInteger> counters = new ConcurrentHashMap<>();
    private final Map<String, Long> timestamps = new ConcurrentHashMap<>();
    
    @Around("@annotation(rateLimit)")
    public Object limit(ProceedingJoinPoint pjp, RateLimit rateLimit) throws Throwable {
        String key = rateLimit.key();
        int limit = rateLimit.limit();
        long interval = rateLimit.interval();
        
        long now = System.currentTimeMillis();
        if (timestamps.getOrDefault(key, 0L) + interval < now) {
            timestamps.put(key, now);
            counters.put(key, new AtomicInteger(0));
        }
        
        if (counters.getOrDefault(key, new AtomicInteger(0)).incrementAndGet() > limit) {
            throw new ResponseStatusException(HttpStatus.TOO_MANY_REQUESTS, "請(qǐng)求過(guò)于頻繁");
        }
        
        return pjp.proceed();
    }
}

// 自定義注解
@Retention(RetentionPolicy.RUNTIME)
@Target(ElementType.METHOD)
public @interface RateLimit {
    String key() default "";
    int limit() default 10;
    long interval() default 1000; // ms
}

// 在Controller方法上使用
@RateLimit(key = "qrcode", limit = 5, interval = 60000)
@GetMapping("/generate")
public void generateQrcode(...) { ... }

6. Apache Commons Pool 對(duì)象池限流

適用于： 需要控制資源并發(fā)使用的場(chǎng)景

GenericObjectPool<Object> pool = new GenericObjectPool<>(new BasePooledObjectFactory<>() {
    @Override
    public Object create() {
        return new Object();
    }
});
pool.setMaxTotal(10); // 最大并發(fā)數(shù)

try {
    Object permit = pool.borrowObject();
    try {
        generateQrcode(...);
    } finally {
        pool.returnObject(permit);
    }
} catch (Exception e) {
    response.sendError(429, "系統(tǒng)繁忙，請(qǐng)稍后再試");
}