如果一个接口的频繁调用会严重影响到系统稳定性的话，那就要考虑限速了。
谷歌有个现有的开源实现：Guava的RateLimiter类，RateLimiter使用的是一种叫令牌桶的流控算法，RateLimiter会按照一定的频率往桶里扔令牌，线程拿到令牌才能执行，比如你希望自己的应用程序QPS不要超过1000，那么RateLimiter设置1000的速率后，就会每秒往桶里扔1000个令牌。 详情请戳：Guava官方文档-RateLimiter类
用法简单说一下吧，首先声明一个限速策略：
final RateLimiter rateLimiter = RateLimiter.create(10.0);然后在需要限流的地方贴上：
rateLimiter.acquire();这样就可以保证每秒钟只能有两次请求成功了。
但是，我需要更多的功能来实现，比如超过限速阈值后也要返回一个东西，这时候就不太好办了，RateLimiter只是在内部采用令牌桶隐式地对接口进行了限速，下面是我的实现：
package com.lovnx.test; import java.util.Timer; import java.util.TimerTask; import java.util.concurrent.ExecutorService; import java.util.concurrent.Executors; import java.util.concurrent.atomic.AtomicInteger; public class mytst { //单位时间内的调用次数 private int flag = 10; //线程数量 private int threadNum = 10; //单位时间1000ms * 60 = 1min private int timeRound = 1000*60; //用来标记调用次数 private AtomicInteger num = new AtomicInteger(0); public static void main(String[] args) { new mytst().call(); } private void call() { ExecutorService executor = Executors.newFixedThreadPool(threadNum); Timer timer = new Timer(); timer.schedule(new TimerTask(){ @Override public void run() { num.set(0);; } }, 0, timeRound); for (int i = 0; i < 1000; i++) { executor.execute(new Runnable() { @Override public void run() { try { Thread.sleep(10000); } catch (InterruptedException e) { e.printStackTrace(); } System.out.println(Thread.currentThread().getName()+" 进来了!"); num.incrementAndGet(); if (num.get() <= flag) { System.out.println(Thread.currentThread().getName()+" 执行任务!"); } else { System.out.println(Thread.currentThread().getName()+" 执行任务失败!调用超限！"); } } }); } executor.shutdown(); } }我比较粗暴，直接用一个原子类AtomicInteger来标记。
在接口层面就是这样的：
package com.lovnx.web; import java.util.Timer; import java.util.TimerTask; import java.util.concurrent.atomic.AtomicInteger; import org.apache.log4j.Logger; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.cloud.client.ServiceInstance; import org.springframework.cloud.client.discovery.DiscoveryClient; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RequestMethod; import org.springframework.web.bind.annotation.RequestParam; import org.springframework.web.bind.annotation.RestController; @RestController public class ComputeController { private final Logger logger = Logger.getLogger(getClass()); //单位时间内的调用次数 private int flag = 10; //单位时间1000ms * 60 = 1min private static int timeRound = 1000*60; //用来标记调用次数 private static AtomicInteger num = new AtomicInteger(0); @Autowired private DiscoveryClient client; static{ Timer timer = new Timer(); timer.schedule(new TimerTask(){ @Override public void run() { num.set(0);; } }, 0, timeRound); } @RequestMapping(value = "/add" ,method = RequestMethod.GET) public String add(@RequestParam Integer a, @RequestParam Integer b) { num.incrementAndGet(); if (num.get() <= flag) { ServiceInstance instance = client.getLocalServiceInstance(); Integer r = a + b; logger.info("/add, host:" + instance.getHost() + ", service_id:" + instance.getServiceId() + ", result:" + r); return "From Service-B, Result is " + r+"\nPort:"+instance.getPort(); } return "调用次数超限，一分钟内最多只能调用10次！"; } }更多内容请戳：代码托管
