JDK1.8并发包之 -- CyclicBarrier

前文讲述CountDownLatch，详见 JDK1.8并发包之 -- CountDownLatch，本文讲述与之类似的并发类--CyclicBarrier，中文叫可循环屏障。

1、第一段，实现目标

 * A synchronization aid that allows a set of threads to all wait for
 * each other to reach a common barrier point.  CyclicBarriers are
 * useful in programs involving a fixed sized party of threads that
 * must occasionally wait for each other. The barrier is called
 * cyclic because it can be re-used after the waiting threads
 * are released.

CyclicBarrier的同步器的目的是让一组多线程相互等待，共同到达一个屏障点，CyclicBarrier经常用在一组固定线程数的场景中，需要互相等待。屏障的作用是让先到达的线程守候在这里，全部线程都到达后，cyclic方法是可循环的，说明可以被重复利用。可重复利用这点是CountDownLatch不具备的。

2、第二段，构造方法

 * A {@code CyclicBarrier} supports an optional {@link Runnable} command
 * that is run once per barrier point, after the last thread in the party
 * arrives, but before any threads are released.
 * This barrier action is useful
 * for updating shared-state before any of the parties continue.

CyclicBarrier的构造方法中能传入一个Runnable对象，最后一个任务线程到达屏障点后，再线程被唤醒前，最后一个线程执行一次Runnable的run方法，barrier非常有用的是在多线程被唤醒前会更新共享状态，这就是可循环利用的原因。

3、第三段、案例

类注释中的代码不是非常清晰，笔者自己写了一段案例。

@Slf4j
public class CyclicBarrierDemo {


    public static void main(String[] args) throws InterruptedException {
        int num = 10;
        CyclicBarrier barrier = new CyclicBarrier(num, new Runnable() {
            @Override
            public void run() {
                log.info("所有年级达到操场，开始升国旗...");
            }
        });

        for (int i=0; i

运行结果如下，显而易见，10个班级都到达操场集合后，才能开始升旗活动。且可重复利用。

15:52:16.321 [Thread-11] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级2 到达操场...
15:52:16.320 [Thread-14] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级5 到达操场...
15:52:16.320 [Thread-4] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级5 到达操场...
15:52:16.321 [Thread-0] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级1 到达操场...
15:52:16.321 [Thread-12] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级3 到达操场...
15:52:16.321 [Thread-16] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级7 到达操场...
15:52:16.321 [Thread-10] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级1 到达操场...
15:52:16.320 [Thread-3] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级4 到达操场...
15:52:16.321 [Thread-2] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级3 到达操场...
15:52:16.321 [Thread-8] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级9 到达操场...
15:52:16.328 [Thread-8] INFO sw.melody.concurrent.CyclicBarrierDemo - 所有年级达到操场，开始升国旗...
15:52:16.322 [Thread-17] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级8 到达操场...
15:52:16.322 [Thread-7] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级8 到达操场...
15:52:16.322 [Thread-6] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级7 到达操场...
15:52:16.322 [Thread-9] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级10 到达操场...
15:52:16.322 [Thread-13] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级4 到达操场...
15:52:16.322 [Thread-1] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级2 到达操场...
15:52:16.322 [Thread-5] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级6 到达操场...
15:52:16.322 [Thread-18] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级9 到达操场...
15:52:16.322 [Thread-15] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级6 到达操场...
15:52:16.322 [Thread-19] INFO sw.melody.concurrent.CyclicBarrierDemo - 班级10 到达操场...
15:52:16.345 [Thread-19] INFO sw.melody.concurrent.CyclicBarrierDemo - 所有年级达到操场，开始升国旗...

4、第四段

 * 
If the barrier action does not rely on the parties being suspended when
 * it is executed, then any of the threads in the party could execute that
 * action when it is released. To facilitate this, each invocation of
 * {@link #await} returns the arrival index of that thread at the barrier.
 * You can then choose which thread should execute the barrier action, for
 * example:
 *  
 {@code
 * if (barrier.await() == 0) {
 *   // log the completion of this iteration
 * }}

这段话的意思是，如果多线程执行的过程中不会暂停互相等待，那么任何被唤醒的线程都可以执行屏障线程的任务。await方法实际上返回的屏障数量的序列号，可以自由判断哪个序列号执行屏障线程任务。如果我们不去重写CyclicBarrier代码，那么这段本身没有意思。

5、第五段，执行异常

 * 
The {@code CyclicBarrier} uses an all-or-none breakage model
 * for failed synchronization attempts: If a thread leaves a barrier
 * point prematurely because of interruption, failure, or timeout, all
 * other threads waiting at that barrier point will also leave
 * abnormally via {@link BrokenBarrierException} (or
 * {@link InterruptedException} if they too were interrupted at about
 * the same time).

CyclicBarrier屏障处理失败，可能由于线程调用interruption方法，执行异常，或者超时，那么等待在屏障点的线程就会抛出BrokenBarrierException。