Java多线程之线程池

线程池处理任务原理

线程池实现参数解析：

public ThreadPoolExecutor(int corePoolSize,
                         int maximumPoolSize,
                         long keepAliveTime,
                         TimeUnit unit,
                         BlockingQueue workQueue) {
    this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue,
         Executors.defaultThreadFactory(), defaultHandler);
}

public ThreadPoolExecutor(int corePoolSize,
                              int maximumPoolSize,
                              long keepAliveTime,
                              TimeUnit unit,
                              BlockingQueue workQueue,
                              ThreadFactory threadFactory,
                              RejectedExecutionHandler handler) {
        if (corePoolSize < 0 ||
            maximumPoolSize <= 0 ||
            maximumPoolSize < corePoolSize ||
            keepAliveTime < 0)
            throw new IllegalArgumentException();
        if (workQueue == null || threadFactory == null || handler == null)
            throw new NullPointerException();
        this.corePoolSize = corePoolSize;
        this.maximumPoolSize = maximumPoolSize;
        this.workQueue = workQueue;
        this.keepAliveTime = unit.toNanos(keepAliveTime);
        this.threadFactory = threadFactory;
        this.handler = handler;
    }

• corePoolSize
  
  • 核心线程数
• maximumPoolSize
  
  • 最大线程数 ，它的值是int java.lang.Integer.MAX_VALUE = 2147483647 [0x7fffffff]
• keepAliveTime
  
  • 线程存活时间
• unit
  
  • 存活时间的时间单位
• workQueue
  
  • 用于装载工作任务的队列
• threadFactory
  
  • 生成现成的线程工厂
• handler
  
  • 饱和策略 RejectedExecutionHandler

BlockingQueue< Runnable>：
- ArrayBlockingQueue：基于数组的有界阻塞队列
- LinkedBlockingQueue： 基于链表的阻塞队列
- SynchronousQueue： 一个没有数据缓冲的BlockingQueue，必须take元素之后才可以put。
- PriorityBlockingQueue： 优先级队列

RejectedExecutionHandler：
ThreadPoolExecutor中有四个默认实现的策略类

1. CallerRunsPolicy： 启用调用者线程执行，除非线程池shut down了，任务就被丢弃了

public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
            if (!e.isShutdown()) {
                r.run();
            }
        }

1. AbortPolicy：直接抛出异常RejectedExecutionException

public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
            throw new RejectedExecutionException("Task " + r.toString() +
                                                 " rejected from " +
                                                 e.toString());
        }

1. DiscardPolicy： 默默的丢弃任务

public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
        }

1. DiscardOldestPolicy：丢弃队列里最久未处理的请求（the oldest unhandled request），然后执行当前任务

public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
            if (!e.isShutdown()) {
                e.getQueue().poll();
                e.execute(r);
            }
        }

小结：java线程池实现是解决资源有限的条件下，尽可能少的因为线程的创建销毁等时间等待和不必要的资源浪费。尽量使用创建好的线程去执行任务，那么来看，他的本质组成就肯定少不了这几个东西：一堆创建好的线程，或者预创建线程数目下可创建的线程，执行的任务，当任务多，但是线程数目有限的时候，对于任务的装载（队列）！以及，当队列满的时候，再有新来的任务的饱和处理策略，比如：是要直接丢弃，还是扔掉队列里面的最远或者最近的任务，执行它还是保存到硬盘或者数据库等等！由此可见，一些线程+队列+饱和处理策略=线程池的实现。

jdk提供的几种线程池

Executors类提供了下述几种方法生成相应的线程池对象：

1. newFixedThreadPool，有这个方法我们一步一步看出来它是核心线程数=最大线程数，工作队列是LinkedBlockingQueue，饱和策略是默认的实现AbortPolicy ，该策略的处理是总是抛出错误。

public static ExecutorService newFixedThreadPool(int nThreads) {
        return new ThreadPoolExecutor(nThreads, nThreads,
                                      0L, TimeUnit.MILLISECONDS,
                                      new LinkedBlockingQueue());
    }

public ThreadPoolExecutor(int corePoolSize,
                              int maximumPoolSize,
                              long keepAliveTime,
                              TimeUnit unit,
                              BlockingQueue workQueue) {
        this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue,
             Executors.defaultThreadFactory(), defaultHandler);
    }

private static final RejectedExecutionHandler defaultHandler =
        new AbortPolicy();

public static class AbortPolicy implements RejectedExecutionHandler {

        public AbortPolicy() { }

        /**
         * Always throws RejectedExecutionException.
         */
        public void rejectedExecution(Runnable r, ThreadPoolExecutor e) {
            throw new RejectedExecutionException("Task " + r.toString() +
                                                 " rejected from " +
                                                 e.toString());
        }
    }

1. newWorkStealingPool jdk1.8提供 暂略
2. newSingleThreadExecutor
   单线程，工作队列是LinkedBlockingQueue，饱和策略是默认的实现AbortPolicy

public static ExecutorService newSingleThreadExecutor() {
        return new FinalizableDelegatedExecutorService
            (new ThreadPoolExecutor(1, 1,
                                    0L, TimeUnit.MILLISECONDS,
                                    new LinkedBlockingQueue()));
    }

1. newCachedThreadPool
   该线程池的特点是初始没有核心线程，执行任务的时候再生成，并且空闲线程的存活时间是1分钟，使用的工作队列是SynchronousQueue这个没有数据缓冲的阻塞队列，饱和策略依然是默认的实现。

public static ExecutorService newCachedThreadPool() {
        return new ThreadPoolExecutor(0, Integer.MAX_VALUE,
                                      60L, TimeUnit.SECONDS,
                                      new SynchronousQueue());
    }

1. newSingleThreadScheduledExecutor
   它调用的是ScheduledThreadPoolExecutor(int corePoolSize)方法，只不过参数是1

public static ScheduledExecutorService newSingleThreadScheduledExecutor() {
        return new DelegatedScheduledExecutorService
            (new ScheduledThreadPoolExecutor(1));
    }

1. newScheduledThreadPool
   DelayedWorkQueue： http://ju.outofmemory.cn/entry/99456

public ScheduledThreadPoolExecutor(int corePoolSize) {
        super(corePoolSize, Integer.MAX_VALUE, 0, NANOSECONDS,
              new DelayedWorkQueue());
    }