线程池

什么是线程池？

池化技术

为什么用线程池?

1 由于设置最大线程数，防止线程过多而导致系统崩溃。

2 线程复用，不需要频繁创建或销毁线程，并且销毁和创建是耗费时间和资源操作，所以提高了效率，节约资源。

线程池原理

核心线程，等待队列，非核心线程，拒绝策略

如果有空闲的线程直接使用，没有空闲的线程并且线程数量未达到corePoolSize，则新建一个线程(核心线程)执行任务

线程数量达到了corePoolSize，则将任务移入队列等待空闲线程将其取出去执行（通过getTask() 方法从阻塞队列中获取等待的任务，如果队列中没有任务，getTask()方法会被阻塞并挂起，不会占用cpu资源，整个getTask操作在自旋下完成）

队列已满，新建线程(非核心线程)执行任务，空闲下来以后，非核心线程会按照时间策略进行销毁

队列已满，总线程数又达到了maximumPoolSize，就会执行任务拒绝策略。

handler：表示当拒绝处理任务时的策略（①AbortPolicy丢弃任务并抛出RejectedExecution异常；②DiscardPolicy丢弃任务，但是不抛出异常；③DiscardOldestPolicy丢弃队列最前面的任务，然后重新尝试执行任务；④CallerRunsPolicy由调用线程处理该任务）

常见四种线程池和自定义线程池

见文档

代码

PoolTest1

public class PoolTest1 {public static void main(String[] args) {
//        ExecutorService executorService1 = Executors.newCachedThreadPool();
//        ExecutorService executorService2 = Executors.newFixedThreadPool(5);
//        ExecutorService executorService3 = Executors.newSingleThreadExecutor();
//        ScheduledExecutorService executorService4 = Executors.newScheduledThreadPool(5);ThreadPoolExecutor executorService = new ThreadPoolExecutor(5, 10, 5, TimeUnit.SECONDS, new LinkedBlockingQueue<>(10), new ThreadPoolExecutor.AbortPolicy());// 执行任务-线程不用我们创建，交给线程池管理，只需要把执行逻辑放进去就OKfor (int i = 0; i < 20; i++) {int finalI = i;executorService.execute(new Runnable() {@Overridepublic void run() {System.out.println("Thread Name： "+Thread.currentThread().getName()+"  I="+ finalI);try {Thread.sleep(5*1000);} catch (InterruptedException e) {e.printStackTrace();}}});}executorService.shutdown();}
}

MyTask

public class MyTask implements Callable<String>  {private Integer integer1;private Integer integer2;public MyTask(Integer integer1,Integer integer2){this.integer1 =integer1;this.integer2 =integer2;}@Overridepublic String call() throws Exception {int i = integer1 + integer2;return "Thread name：" + Thread.currentThread().getName()+"  i="+i;}
}

PoolTest2

public class PoolTest2 {public static void main(String[] args) throws ExecutionException, InterruptedException {
//        ExecutorService executorService1 = Executors.newCachedThreadPool();
//        ExecutorService executorService2 = Executors.newFixedThreadPool(5);
//        ExecutorService executorService3 = Executors.newSingleThreadExecutor();
//        ScheduledExecutorService executorService4 = Executors.newScheduledThreadPool(5);ThreadPoolExecutor executorService = new ThreadPoolExecutor(5, 10, 5, TimeUnit.SECONDS, new LinkedBlockingQueue<>(10), new ThreadPoolExecutor.AbortPolicy());// 执行任务-线程不用我们创建，交给线程池管理，只需要把执行逻辑放进去就OKArrayList<Future<String>> futures = new ArrayList<>();for (int i = 0; i < 20; i++) {MyTask myTask = new MyTask(3,4);Future<String> future = executorService.submit(myTask);// future.get()为myTask.call()的返回值futures.add(future);}List<String> resutlts = new ArrayList<>();for (int i = 0; i < futures.size(); i++) {String resutlt = futures.get(i).get();resutlts.add(resutlt);}System.out.println("===="+resutlts.get(19));executorService.shutdown();}
}