Java线程池参数详解

发布时间：2023-05-18

一、核心参数

  1. corePoolSize 核心池的大小，即线程池维护线程的最少数量。当提交一个任务到线程池时，线程池会创建一个线程来执行任务，即使其他线程处于空闲状态。如果池中的线程数达到 corePoolSize 后，新提交的任务将被放入队列中等待。如果队列已满，且池中的线程数小于 maximumPoolSize，则创建新的线程执行任务。如果池中的线程数达到 maximumPoolSize，则根据设置的拒绝策略来处理任务。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>());
  2. maximumPoolSize 线程池最大数量。当线程池中的线程数目达到这个数目时，新的任务会被放入队列中等待。如果队列满了，调用者就必须自己处理这个任务。如果使用无界的任务队列这个参数就没有任何作用。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>());
  3. keepAliveTime 线程池中工作线程空闲后，保持存活的时间。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>());

二、任务队列参数

  1. workQueue 任务队列，用来存储等待执行的任务。常用的有四种队列： 
    // 线程池中任务队列为SynchronousQueue
ThreadPoolExecutor threadPoolExecutor1 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new SynchronousQueue<Runnable>());
// 线程池中任务队列为LinkedBlockingQueue
ThreadPoolExecutor threadPoolExecutor2 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>());
// 线程池中任务队列为ArrayBlockingQueue
ThreadPoolExecutor threadPoolExecutor3 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new ArrayBlockingQueue<Runnable>(100));
// 线程池中任务队列为PriorityBlockingQueue
ThreadPoolExecutor threadPoolExecutor4 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new PriorityBlockingQueue<Runnable>());
  2. capacity 任务队列的容量。如果任务数量达到这个值，新提交的任务将会被拒绝，并根据设置的拒绝策略来处理任务。 
    // 容量为100的有界队列
ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new ArrayBlockingQueue<Runnable>(100));

三、超时策略参数

  1. allowCoreThreadTimeOut 该属性为 true 时，表示允许为核心线程设置超时时间，即当线程数大于核心线程数时，核心线程也会参与超时检查。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>());
threadPoolExecutor.allowCoreThreadTimeOut(true);
  2. rejectedExecutionHandler 当线程池中的资源已经全部使用，添加新线程被拒绝时，会进行拒绝处理。常用的拒绝策略有四种： 
    // 拒绝策略为AbortPolicy
ThreadPoolExecutor threadPoolExecutor1 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadPoolExecutor.AbortPolicy());
// 拒绝策略为DiscardPolicy
ThreadPoolExecutor threadPoolExecutor2 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadPoolExecutor.DiscardPolicy());
// 拒绝策略为DiscardOldestPolicy
ThreadPoolExecutor threadPoolExecutor3 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadPoolExecutor.DiscardOldestPolicy());
// 拒绝策略为CallerRunsPolicy
ThreadPoolExecutor threadPoolExecutor4 = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime,
            TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadPoolExecutor.CallerRunsPolicy());

四、线程池扩展参数

  1. ThreadFactory 通过 ThreadFactory 方式自定义线程池中的线程对象，可以自定义线程的名字、是否为守护线程等。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadFactory() {
   private final AtomicInteger atomicInteger = new AtomicInteger(0);
   public Thread newThread(Runnable r) {
       Thread thread = new Thread(r);
       thread.setName("Thread-" + atomicInteger.getAndIncrement());
       return thread;
   }
});
  2. beforeExecute 和 afterExecute 在每个任务之前和之后执行的操作。可以用来统计任务的执行时间等。 
    ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit.SECONDS, new LinkedBlockingQueue<Runnable>(), new ThreadPoolExecutor.AbortPolicy());
threadPoolExecutor.setThreadFactory(new ThreadFactory() {
   private final AtomicInteger atomicInteger = new AtomicInteger(0);
   public Thread newThread(Runnable r) {
       Thread thread = new Thread(r);
       thread.setName("Thread-" + atomicInteger.getAndIncrement());
       return thread;
   }
});
// 执行任务前处理
threadPoolExecutor.prestartAllCoreThreads();
threadPoolExecutor.execute(() -> {
    // 任务执行时间统计
    long startTime = System.currentTimeMillis();
    try {
        // 执行任务
    } finally {
        // 任务执行时间统计
        long endTime = System.currentTimeMillis();
        log.info("任务执行时间: " + (endTime - startTime) + " ms");
    }
});

五、总结

本文主要对 Java 线程池中的主要参数进行了详细的阐述，包括核心参数、任务队列参数、超时策略参数以及线程池扩展参数。在使用线程池时，需要根据具体业务场景选择合适的参数配置，避免线程池资源浪费或者任务无法及时执行的情况。

推荐文章