Java Redis多限流-编程知识

Java Redis多限流

news/2025/1/5 15:42:18/文章来源:https://www.cnblogs.com/mybook000/p/18649909

在现代Web应用中，限流（Rate Limiting）是保护系统资源和防止滥用的重要机制。Redis由于其高性能和原子操作特性，成为实现限流的理想选择。本文将详细介绍如何在Java中使用Redis实现多种限流策略，包括固定窗口限流、滑动窗口限流和令牌桶算法。

一、准备工作

1. 安装Redis

确保Redis已经安装并正在运行。可以通过以下命令安装Redis：

sudo apt-get update
sudo apt-get install redis-server
sudo service redis-server start

2. 添加依赖

在Java项目中使用Redis，推荐使用Jedis库。以下是Maven依赖：

<dependency><groupId>redis.clients</groupId><artifactId>jedis</artifactId><version>3.6.0</version>
</dependency>

二、固定窗口限流

固定窗口限流是最简单的限流算法。它将时间划分为固定的窗口，每个窗口内限制请求次数。

1. 实现逻辑

每当一个请求到达时，检查当前窗口内的请求数量是否超过限制。如果未超过，则允许请求并增加计数；否则，拒绝请求。

2. 示例代码

import redis.clients.jedis.Jedis;public class FixedWindowRateLimiter {private static final String REDIS_HOST = "localhost";private static final int REDIS_PORT = 6379;private static final int LIMIT = 10; // 每窗口的最大请求数private static final int WINDOW_SIZE = 60; // 窗口大小（秒）private Jedis jedis;public FixedWindowRateLimiter() {this.jedis = new Jedis(REDIS_HOST, REDIS_PORT);}public boolean isAllowed(String userId) {String key = "rate_limit:" + userId;long currentWindow = System.currentTimeMillis() / 1000 / WINDOW_SIZE;String windowKey = key + ":" + currentWindow;if (jedis.exists(windowKey)) {if (Integer.parseInt(jedis.get(windowKey)) < LIMIT) {jedis.incr(windowKey);return true;} else {return false;}} else {jedis.setex(windowKey, WINDOW_SIZE, "1");return true;}}public static void main(String[] args) {FixedWindowRateLimiter limiter = new FixedWindowRateLimiter();String userId = "user123";for (int i = 0; i < 15; i++) {System.out.println("Request " + (i + 1) + ": " + limiter.isAllowed(userId));}}
}

三、滑动窗口限流

滑动窗口限流能够更精确地控制请求速率，避免固定窗口算法中临界点的突发流量问题。

1. 实现逻辑

滑动窗口限流记录每个请求的时间戳，并在每次请求时清理过期的记录。

2. 示例代码

import redis.clients.jedis.Jedis;
import java.util.stream.Collectors;
import java.util.List;public class SlidingWindowRateLimiter {private static final String REDIS_HOST = "localhost";private static final int REDIS_PORT = 6379;private static final int LIMIT = 10; // 每窗口的最大请求数private static final int WINDOW_SIZE = 60; // 窗口大小（秒）private Jedis jedis;public SlidingWindowRateLimiter() {this.jedis = new Jedis(REDIS_HOST, REDIS_PORT);}public boolean isAllowed(String userId) {String key = "sliding_rate_limit:" + userId;long currentTime = System.currentTimeMillis() / 1000;long windowStart = currentTime - WINDOW_SIZE;// 清理过期请求jedis.zremrangeByScore(key, 0, windowStart);// 获取当前窗口内的请求数量long count = jedis.zcard(key);if (count < LIMIT) {jedis.zadd(key, currentTime, String.valueOf(currentTime));jedis.expire(key, WINDOW_SIZE);return true;} else {return false;}}public static void main(String[] args) {SlidingWindowRateLimiter limiter = new SlidingWindowRateLimiter();String userId = "user123";for (int i = 0; i < 15; i++) {System.out.println("Request " + (i + 1) + ": " + limiter.isAllowed(userId));}}
}

四、令牌桶算法

令牌桶算法是一种常用的流量整形算法，能够控制数据的流入速率。

1. 实现逻辑

令牌桶算法通过定时向桶中添加令牌，每次请求消耗一个令牌，如果桶为空则拒绝请求。

2. 示例代码

import redis.clients.jedis.Jedis;public class TokenBucketRateLimiter {private static final String REDIS_HOST = "localhost";private static final int REDIS_PORT = 6379;private static final int MAX_TOKENS = 10; // 最大令牌数private static final int REFILL_RATE = 1; // 每秒添加令牌数private Jedis jedis;public TokenBucketRateLimiter() {this.jedis = new Jedis(REDIS_HOST, REDIS_PORT);}public boolean isAllowed(String userId) {String key = "token_bucket:" + userId;long currentTime = System.currentTimeMillis() / 1000;// 获取最后更新时间和当前令牌数String[] bucketInfo = jedis.hmget(key, "tokens", "timestamp").toArray(new String[0]);int tokens = bucketInfo[0] != null ? Integer.parseInt(bucketInfo[0]) : MAX_TOKENS;long lastRefillTime = bucketInfo[1] != null ? Long.parseLong(bucketInfo[1]) : currentTime;// 计算需要添加的令牌数long timeDiff = currentTime - lastRefillTime;int newTokens = (int) Math.min(MAX_TOKENS, tokens + timeDiff * REFILL_RATE);if (newTokens > 0) {jedis.hmset(key, Map.of("tokens", String.valueOf(newTokens - 1), "timestamp", String.valueOf(currentTime)));jedis.expire(key, MAX_TOKENS / REFILL_RATE); // 设置过期时间return true;} else {return false;}}public static void main(String[] args) {TokenBucketRateLimiter limiter = new TokenBucketRateLimiter();String userId = "user123";for (int i = 0; i < 15; i++) {System.out.println("Request " + (i + 1) + ": " + limiter.isAllowed(userId));}}
}