java.util.concurrent 多线程框架

（来源于http://www.zhuaxia.com/item/590227619/）

JDK5中的一个亮点就是将Doug Lea的并发库引入到Java标准库中。Doug Lea确实是一个牛人，能教书，能出书，能编码，不过这在国外还是比较普遍的，而国内的教授们就相差太远了。

一般的服务器都需要线程池，比如Web、FTP等服务器，不过它们一般都自己实现了线程池，比如以前介绍过的Tomcat、Resin和 Jetty等，现在有了JDK5，我们就没有必要重复造车轮了，直接使用就可以，何况使用也很方便，性能也非常高。

 package  concurrent;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 public   class  TestThreadPool {
 public   static   void  main(String args[])  throws  InterruptedException {
 // only two threads
 ExecutorService exec = Executors.newFixedThreadPool(2 );
 for ( int  index =  0 ; index <  100 ; index++) {
 Runnable run = new  Runnable() {
 public   void  run() {
 long  time = ( long ) (Math.random() *  1000 );
 System.out.println(“Sleeping ” + time + “ms”);
 try  {
 Thread.sleep(time);
 } catch  (InterruptedException e) {
 }
 }
 };
 exec.execute(run);
 }
 // must shutdown
 exec.shutdown();
 }
 }

package concurrent;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
public class TestThreadPool {
public static void main(String args[]) throws InterruptedException {
// only two threads
ExecutorService exec = Executors.newFixedThreadPool(2);
for(int index = 0; index < 100; index++) {
Runnable run = new Runnable() {
public void run() {
long time = (long) (Math.random() * 1000);
System.out.println(“Sleeping ” + time + “ms”);
try {
Thread.sleep(time);
} catch (InterruptedException e) {
}
}
};
exec.execute(run);
}
// must shutdown
exec.shutdown();
}
}

上面是一个简单的例子，使用了2个大小的线程池来处理100个线程。但有一个问题：在for循环的过程中，会等待线程池有空闲的线程，所以主线程 会阻塞的。为了解决这个问题，一般启动一个线程来做for循环，就是为了避免由于线程池满了造成主线程阻塞。不过在这里我没有这样处理。[重要修正：经过 测试，即使线程池大小小于实际线程数大小，线程池也不会阻塞的，这与Tomcat的线程池不同，它将Runnable实例放到一个“无限”的 BlockingQueue中，所以就不用一个线程启动for循环，Doug Lea果然厉害]

另外它使用了Executors的静态函数生成一个固定的线程池，顾名思义，线程池的线程是不会释放的，即使它是Idle。这就会产生性能问题， 比如如果线程池的大小为200，当全部使用完毕后，所有的线程会继续留在池中，相应的内存和线程切换（while(true)+sleep循环）都会增 加。如果要避免这个问题，就必须直接使用ThreadPoolExecutor()来构造。可以像Tomcat的线程池一样设置“最大线程数”、“最小线 程数”和“空闲线程keepAlive的时间”。通过这些可以基本上替换Tomcat的线程池实现方案。

需要注意的是线程池必须使用shutdown来显式关闭，否则主线程就无法退出。shutdown也不会阻塞主线程。

许多长时间运行的应用有时候需要定时运行任务完成一些诸如统计、优化等工作，比如在电信行业中处理用户话单时，需要每隔1分钟处理话单；网站每天 凌晨统计用户访问量、用户数；大型超时凌晨3点统计当天销售额、以及最热卖的商品；每周日进行数据库备份；公司每个月的10号计算工资并进行转帐等，这些 都是定时任务。通过 java的并发库concurrent可以轻松的完成这些任务，而且非常的简单。

 package  concurrent;
 import   static  java.util.concurrent.TimeUnit.SECONDS;
 import  java.util.Date;
 import  java.util.concurrent.Executors;
 import  java.util.concurrent.ScheduledExecutorService;
 import  java.util.concurrent.ScheduledFuture;
 public   class  TestScheduledThread {
 public   static   void  main(String[] args) {
 final  ScheduledExecutorService scheduler = Executors
 .newScheduledThreadPool(2 );
 final  Runnable beeper =  new  Runnable() {
 int  count =  0 ;
 public   void  run() {
 System.out.println(new  Date() + ” beep ” + (++count));
 }
 };
 // 1秒钟后运行，并每隔2秒运行一次
 final  ScheduledFuture beeperHandle = scheduler.scheduleAtFixedRate(
 beeper, 1 ,  2 , SECONDS);
 // 2秒钟后运行，并每次在上次任务运行完后等待5秒后重新运行
 final  ScheduledFuture beeperHandle2 = scheduler
 .scheduleWithFixedDelay(beeper, 2 ,  5 , SECONDS);
 // 30秒后结束关闭任务，并且关闭Scheduler
 scheduler.schedule(new  Runnable() {
 public   void  run() {
 beeperHandle.cancel(true );
 beeperHandle2.cancel(true );
 scheduler.shutdown();
 }
 }, 30 , SECONDS);
 }
 }

package concurrent;
import static java.util.concurrent.TimeUnit.SECONDS;
import java.util.Date;
import java.util.concurrent.Executors;
import java.util.concurrent.ScheduledExecutorService;
import java.util.concurrent.ScheduledFuture;
public class TestScheduledThread {
public static void main(String[] args) {
final ScheduledExecutorService scheduler = Executors
.newScheduledThreadPool(2);
final Runnable beeper = new Runnable() {
int count = 0;
public void run() {
System.out.println(new Date() + ” beep ” + (++count));
}
};
// 1秒钟后运行，并每隔2秒运行一次
final ScheduledFuture beeperHandle = scheduler.scheduleAtFixedRate(
beeper, 1, 2, SECONDS);
// 2秒钟后运行，并每次在上次任务运行完后等待5秒后重新运行
final ScheduledFuture beeperHandle2 = scheduler
.scheduleWithFixedDelay(beeper, 2, 5, SECONDS);
// 30秒后结束关闭任务，并且关闭Scheduler
scheduler.schedule(new Runnable() {
public void run() {
beeperHandle.cancel(true);
beeperHandle2.cancel(true);
scheduler.shutdown();
}
}, 30, SECONDS);
}
}

为了退出进程，上面的代码中加入了关闭Scheduler的操作。而对于24小时运行的应用而言，是没有必要关闭Scheduler的。

在实际应用中，有时候需要多个线程同时工作以完成同一件事情，而且在完成过程中，往往会等待其他线程都完成某一阶段后再执行，等所有线程都到达某 一个阶段后再统一执行。

比如有几个旅行团需要途经深圳、广州、韶关、长沙最后到达武汉。旅行团中有自驾游的，有徒步的，有乘坐旅游大巴的；这些旅行团同时出发，并且每到 一个目的地，都要等待其他旅行团到达此地后再同时出发，直到都到达终点站武汉。

这时候CyclicBarrier就可以派上用场。CyclicBarrier最重要的属性就是参与者个数，另外最要方法是await()。当所 有线程都调用了await()后，就表示这些线程都可以继续执行，否则就会等待。

 package  concurrent;
 import  java.text.SimpleDateFormat;
 import  java.util.Date;
 import  java.util.concurrent.BrokenBarrierException;
 import  java.util.concurrent.CyclicBarrier;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 public   class  TestCyclicBarrier {
 // 徒步需要的时 间: Shenzhen, Guangzhou, Shaoguan, Changsha, Wuhan
 private   static   int [] timeWalk = {  5 ,  8 ,  15 ,  15 ,  10  };
 // 自驾游
 private   static   int [] timeSelf = {  1 ,  3 ,  4 ,  4 ,  5  };
 // 旅游大巴
 private   static   int [] timeBus = {  2 ,  4 ,  6 ,  6 ,  7  };

 static  String now() {
 SimpleDateFormat sdf = new  SimpleDateFormat(“HH:mm:ss”);
 return  sdf.format( new  Date()) + “: “;
 }

 static   class  Tour  implements  Runnable {
 private   int [] times;
 private  CyclicBarrier barrier;
 private  String tourName;
 public  Tour(CyclicBarrier barrier, String tourName,  int [] times) {
 this .times = times;
 this .tourName = tourName;
 this .barrier = barrier;
 }
 public   void  run() {
 try  {
 Thread.sleep(times[0 ] *  1000 );
 System.out.println(now() + tourName + ” Reached Shenzhen”);
 barrier.await();
 Thread.sleep(times[1 ] *  1000 );
 System.out.println(now() + tourName + ” Reached Guangzhou”);
 barrier.await();
 Thread.sleep(times[2 ] *  1000 );
 System.out.println(now() + tourName + ” Reached Shaoguan”);
 barrier.await();
 Thread.sleep(times[3 ] *  1000 );
 System.out.println(now() + tourName + ” Reached Changsha”);
 barrier.await();
 Thread.sleep(times[4 ] *  1000 );
 System.out.println(now() + tourName + ” Reached Wuhan”);
 barrier.await();
 } catch  (InterruptedException e) {
 } catch  (BrokenBarrierException e) {
 }
 }
 }

 public   static   void  main(String[] args) {
 // 三个旅行团
 CyclicBarrier barrier = new  CyclicBarrier( 3 );
 ExecutorService exec = Executors.newFixedThreadPool(3 );
 exec.submit(new  Tour(barrier, “WalkTour”, timeWalk));
 exec.submit(new  Tour(barrier, “SelfTour”, timeSelf));
 exec.submit(new  Tour(barrier, “BusTour”, timeBus));
 exec.shutdown();
 }
 }

package concurrent;
import java.text.SimpleDateFormat;
import java.util.Date;
import java.util.concurrent.BrokenBarrierException;
import java.util.concurrent.CyclicBarrier;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
public class TestCyclicBarrier {
// 徒步需要的时间: Shenzhen, Guangzhou, Shaoguan, Changsha, Wuhan
private static int[] timeWalk = { 5, 8, 15, 15, 10 };
// 自驾游
private static int[] timeSelf = { 1, 3, 4, 4, 5 };
// 旅游大巴
private static int[] timeBus = { 2, 4, 6, 6, 7 };

static String now() {
SimpleDateFormat sdf = new SimpleDateFormat(“HH:mm:ss”);
return sdf.format(new Date()) + “: “;
}

static class Tour implements Runnable {
private int[] times;
private CyclicBarrier barrier;
private String tourName;
public Tour(CyclicBarrier barrier, String tourName, int[] times) {
this.times = times;
this.tourName = tourName;
this.barrier = barrier;
}
public void run() {
try {
Thread.sleep(times[0] * 1000);
System.out.println(now() + tourName + ” Reached Shenzhen”);
barrier.await();
Thread.sleep(times[1] * 1000);
System.out.println(now() + tourName + ” Reached Guangzhou”);
barrier.await();
Thread.sleep(times[2] * 1000);
System.out.println(now() + tourName + ” Reached Shaoguan”);
barrier.await();
Thread.sleep(times[3] * 1000);
System.out.println(now() + tourName + ” Reached Changsha”);
barrier.await();
Thread.sleep(times[4] * 1000);
System.out.println(now() + tourName + ” Reached Wuhan”);
barrier.await();
} catch (InterruptedException e) {
} catch (BrokenBarrierException e) {
}
}
}

public static void main(String[] args) {
// 三个旅行团
CyclicBarrier barrier = new CyclicBarrier(3);
ExecutorService exec = Executors.newFixedThreadPool(3);
exec.submit(new Tour(barrier, “WalkTour”, timeWalk));
exec.submit(new Tour(barrier, “SelfTour”, timeSelf));
exec.submit(new Tour(barrier, “BusTour”, timeBus));
exec.shutdown();
}
}

运行结果：
00:02:25: SelfTour Reached Shenzhen
00:02:25: BusTour Reached Shenzhen
00:02:27: WalkTour Reached Shenzhen
00:02:30: SelfTour Reached Guangzhou
00:02:31: BusTour Reached Guangzhou
00:02:35: WalkTour Reached Guangzhou
00:02:39: SelfTour Reached Shaoguan
00:02:41: BusTour Reached Shaoguan

并发库中的BlockingQueue是一个比较好玩的类，顾名思义，就是阻塞队列。该类主要提供了两个方法put()和take()，前者将一 个对象放到队列中，如果队列已经满了，就等待直到有空闲节点；后者从head取一个对象，如果没有对象，就等待直到有可取的对象。

下面的例子比较简单，一个读线程，用于将要处理的文件对象添加到阻塞队列中，另外四个写线程用于取出文件对象，为了模拟写操作耗时长的特点，特让 线程睡眠一段随机长度的时间。另外，该Demo也使用到了线程池和原子整型（AtomicInteger），AtomicInteger可以在并发情况下 达到原子化更新，避免使用了synchronized，而且性能非常高。由于阻塞队列的put和take操作会阻塞，为了使线程退出，特在队列中添加了一 个“标识”，算法中也叫“哨兵”，当发现这个哨兵后，写线程就退出。

当然线程池也要显式退出了。

 package  concurrent;
 import  java.io.File;
 import  java.io.FileFilter;
 import  java.util.concurrent.BlockingQueue;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 import  java.util.concurrent.LinkedBlockingQueue;
 import  java.util.concurrent.atomic.AtomicInteger;

 public   class  TestBlockingQueue {
 static   long  randomTime() {
 return  ( long ) (Math.random() *  1000 );
 }

 public   static   void  main(String[] args) {
 // 能容纳100个文件
 final  BlockingQueue queue =  new  LinkedBlockingQueue( 100 );
 // 线程池
 final  ExecutorService exec = Executors.newFixedThreadPool( 5 );
 final  File root =  new  File(“F:\\JavaLib”);
 // 完成标志
 final  File exitFile =  new  File(“”);
 // 读个数
 final  AtomicInteger rc =  new  AtomicInteger();
 // 写个数
 final  AtomicInteger wc =  new  AtomicInteger();
 // 读线程
 Runnable read = new  Runnable() {
 public   void  run() {
 scanFile(root);
 scanFile(exitFile);
 }

 public   void  scanFile(File file) {
 if  (file.isDirectory()) {
 File[] files = file.listFiles(new  FileFilter() {
 public   boolean  accept(File pathname) {
 return  pathname.isDirectory()
 || pathname.getPath().endsWith(“.java”);
 }
 });
 for  (File one : files)
 scanFile(one);
 } else  {
 try  {
 int  index = rc.incrementAndGet();
 System.out.println(“Read0: ” + index + ” “
 + file.getPath());
 queue.put(file);
 } catch  (InterruptedException e) {
 }
 }
 }
 };
 exec.submit(read);
 // 四个写线程
 for  ( int  index =  0 ; index <  4 ; index++) {
 // write thread
 final   int  NO = index;
 Runnable write = new  Runnable() {
 String threadName = “Write” + NO;
 public   void  run() {
 while  ( true ) {
 try  {
 Thread.sleep(randomTime());
 int  index = wc.incrementAndGet();
 File file = queue.take();
 // 队列已经无对象
 if  (file == exitFile) {
 // 再次添加”标志”，以让其他线程正常退出
 queue.put(exitFile);
 break ;
 }
 System.out.println(threadName + “: ” + index + ” “
 + file.getPath());
 } catch  (InterruptedException e) {
 }
 }
 }
 };
 exec.submit(write);
 }
 exec.shutdown();
 }
 }

package concurrent;
import java.io.File;
import java.io.FileFilter;
import java.util.concurrent.BlockingQueue;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.LinkedBlockingQueue;
import java.util.concurrent.atomic.AtomicInteger;

public class TestBlockingQueue {
static long randomTime() {
return (long) (Math.random() * 1000);
}

public static void main(String[] args) {
// 能容纳100个文件
final BlockingQueue queue = new LinkedBlockingQueue(100);
// 线程池
final ExecutorService exec = Executors.newFixedThreadPool(5);
final File root = new File(“F:\\JavaLib”);
// 完成标志
final File exitFile = new File(“”);
// 读个数
final AtomicInteger rc = new AtomicInteger();
// 写个数
final AtomicInteger wc = new AtomicInteger();
// 读线程
Runnable read = new Runnable() {
public void run() {
scanFile(root);
scanFile(exitFile);
}

public void scanFile(File file) {
if (file.isDirectory()) {
File[] files = file.listFiles(new FileFilter() {
public boolean accept(File pathname) {
return pathname.isDirectory()
|| pathname.getPath().endsWith(“.java”);
}
});
for (File one : files)
scanFile(one);
} else {
try {
int index = rc.incrementAndGet();
System.out.println(“Read0: ” + index + ” “
+ file.getPath());
queue.put(file);
} catch (InterruptedException e) {
}
}
}
};
exec.submit(read);
// 四个写线程
for (int index = 0; index < 4; index++) {
// write thread
final int NO = index;
Runnable write = new Runnable() {
String threadName = “Write” + NO;
public void run() {
while (true) {
try {
Thread.sleep(randomTime());
int index = wc.incrementAndGet();
File file = queue.take();
// 队列已经无对象
if (file == exitFile) {
// 再次添加”标志”，以让其他线程正常退出
queue.put(exitFile);
break;
}
System.out.println(threadName + “: ” + index + ” “
+ file.getPath());
} catch (InterruptedException e) {
}
}
}
};
exec.submit(write);
}
exec.shutdown();
}
}

从名字可以看出，CountDownLatch是一个倒数计数的锁，当倒数到0时触发事件，也就是开锁，其他人就可以进入了。在一些应用场合中， 需要等待某个条件达到要求后才能做后面的事情；同时当线程都完成后也会触发事件，以便进行后面的操作。


CountDownLatch最重要的方法是countDown()和await()，前者主要是倒数一次，后者是等待倒数到0，如果没有到达 0，就只有阻塞等待了。

一个CountDouwnLatch实例是不能重复使用的，也就是说它是一次性的，锁一经被打开就不能再关闭使用了，如果想重复使用，请考虑使用 CyclicBarrier。

下面的例子简单的说明了CountDownLatch的使用方法，模拟了100米赛跑，10名选手已经准备就绪，只等裁判一声令下。当所有人都到 达终点时，比赛结束。

同样，线程池需要显式shutdown。

 package  concurrent;

 import  java.util.concurrent.CountDownLatch;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;

 public   class  TestCountDownLatch {
 public   static   void  main(String[] args)  throws  InterruptedException {
 // 开始的倒数锁
 final  CountDownLatch begin =  new  CountDownLatch( 1 );
 // 结束的倒数锁
 final  CountDownLatch end =  new  CountDownLatch( 10 );
 // 十名选手
 final  ExecutorService exec = Executors.newFixedThreadPool( 10 );
 for ( int  index =  0 ; index <  10 ; index++) {
 final   int  NO = index +  1 ;
 Runnable run = new  Runnable(){
 public   void  run() {
 try  {
 begin.await();
 Thread.sleep((long ) (Math.random() *  10000 ));
 System.out.println(“No.” + NO + ” arrived”);
 } catch  (InterruptedException e) {
 } finally  {
 end.countDown();
 }
 }
 };
 exec.submit(run);
 }
 System.out.println(“Game Start”);
 begin.countDown();
 end.await();
 System.out.println(“Game Over”);
 exec.shutdown();
 }
 }

package concurrent;

import java.util.concurrent.CountDownLatch;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

public class TestCountDownLatch {
public static void main(String[] args) throws InterruptedException {
// 开始的倒数锁
final CountDownLatch begin = new CountDownLatch(1);
// 结束的倒数锁
final CountDownLatch end = new CountDownLatch(10);
// 十名选手
final ExecutorService exec = Executors.newFixedThreadPool(10);
for(int index = 0; index < 10; index++) {
final int NO = index + 1;
Runnable run = new Runnable(){
public void run() {
try {
begin.await();
Thread.sleep((long) (Math.random() * 10000));
System.out.println(“No.” + NO + ” arrived”);
} catch (InterruptedException e) {
} finally {
end.countDown();
}
}
};
exec.submit(run);
}
System.out.println(“Game Start”);
begin.countDown();
end.await();
System.out.println(“Game Over”);
exec.shutdown();
}
}

运行结果:
Game Start
No.4 arrived
No.1 arrived
No.7 arrived
No.9 arrived
No.3 arrived
No.2 arrived
No.8 arrived
No.10 arrived
No.6 arrived
No.5 arrived
Game Over

有时候在实际应用中，某些操作很耗时，但又不是不可或缺的步骤。比如用网页浏览器浏览新闻时，最重要的是要显示文字内容，至于与新闻相匹配的图片 就没有那么重要的，所以此时首先保证文字信息先显示，而图片信息会后显示，但又不能不显示，由于下载图片是一个耗时的操作，所以必须一开始就得下载。


Java的并发库的Future类就可以满足这个要求。Future的重要方法包括get()和cancel()，get()获取数据对象，如果 数据没有加载，就会阻塞直到取到数据，而 cancel()是取消数据加载。另外一个get(timeout)操作，表示如果在timeout时间内没有取到就失败返回，而不再阻塞。

下面的Demo简单的说明了Future的使用方法：一个非常耗时的操作必须一开始启动，但又不能一直等待；其他重要的事情又必须做，等完成后， 就可以做不重要的事情。

 package  concurrent;

 import  java.util.concurrent.Callable;
 import  java.util.concurrent.ExecutionException;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 import  java.util.concurrent.Future;

 public   class  TestFutureTask {
 public   static   void  main(String[] args) throws  InterruptedException,
 ExecutionException {
 final  ExecutorService exec = Executors.newFixedThreadPool( 5 );
 Callable call = new  Callable() {
 public  String call()  throws  Exception {
 Thread.sleep(1000  *  5 );
 return  “Other less important but longtime things.”;
 }
 };
 Future task = exec.submit(call);
 // 重要的事情
 Thread.sleep(1000  *  3 );
 System.out.println(“Let’s do  important things.”);
 // 其他不重要的事情
 String obj = task.get();
 System.out.println(obj);
 // 关闭线程池
 exec.shutdown();
 }
 }

package concurrent;

import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

public class TestFutureTask {
public static void main(String[] args)throws InterruptedException,
ExecutionException {
final ExecutorService exec = Executors.newFixedThreadPool(5);
Callable call = new Callable() {
public String call() throws Exception {
Thread.sleep(1000 * 5);
return “Other less important but longtime things.”;
}
};
Future task = exec.submit(call);
// 重要的事情
Thread.sleep(1000 * 3);
System.out.println(“Let’s do important things.”);
// 其他不重要的事情
String obj = task.get();
System.out.println(obj);
// 关闭线程池
exec.shutdown();
}
}

运行结果：
Let’s do important things.
Other less important but longtime things.

考虑以下场景：浏览网页时，浏览器了5个线程下载网页中的图片文件，由于图片大小、网站访问速度等诸多因素的影响，完成图片下载的时间就会有很大 的不同。如果先下载完成的图片就会被先显示到界面上，反之，后下载的图片就后显示。


Java的并发库的CompletionService可以满足这种场景要求。该接口有两个重要方法：submit()和take()。 submit用于提交一个runnable或者callable，一般会提交给一个线程池处理；而take就是取出已经执行完毕runnable或者 callable实例的Future对象，如果没有满足要求的，就等待了。 CompletionService还有一个对应的方法poll，该方法与take类似，只是不会等待，如果没有满足要求，就返回null对象。

 package  concurrent;

 import  java.util.concurrent.Callable;
 import  java.util.concurrent.CompletionService;
 import  java.util.concurrent.ExecutionException;
 import  java.util.concurrent.ExecutorCompletionService;
 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 import  java.util.concurrent.Future;

 public   class  TestCompletionService {
 public   static   void  main(String[] args)  throws  InterruptedException,
 ExecutionException {
 ExecutorService exec = Executors.newFixedThreadPool(10 );
 CompletionService serv =
 new  ExecutorCompletionService(exec);

 for  ( int  index =  0 ; index <  5 ; index++) {
 final   int  NO = index;
 Callable downImg = new  Callable() {
 public  String call()  throws  Exception {
 Thread.sleep((long ) (Math.random() *  10000 ));
 return  “Downloaded Image ” + NO;
 }
 };
 serv.submit(downImg);
 }

 Thread.sleep(1000  *  2 );
 System.out.println(“Show web content”);
 for  ( int  index =  0 ; index <  5 ; index++) {
 Future task = serv.take();
 String img = task.get();
 System.out.println(img);
 }
 System.out.println(“End”);
 // 关闭线程池
 exec.shutdown();
 }
 }

package concurrent;

import java.util.concurrent.Callable;
import java.util.concurrent.CompletionService;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorCompletionService;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

public class TestCompletionService {
public static void main(String[] args) throws InterruptedException,
ExecutionException {
ExecutorService exec = Executors.newFixedThreadPool(10);
CompletionService serv =
new ExecutorCompletionService(exec);

for (int index = 0; index < 5; index++) {
final int NO = index;
Callable downImg = new Callable() {
public String call() throws Exception {
Thread.sleep((long) (Math.random() * 10000));
return “Downloaded Image ” + NO;
}
};
serv.submit(downImg);
}

Thread.sleep(1000 * 2);
System.out.println(“Show web content”);
for (int index = 0; index < 5; index++) {
Future task = serv.take();
String img = task.get();
System.out.println(img);
}
System.out.println(“End”);
// 关闭线程池
exec.shutdown();
}
}

运行结果:
Show web content
Downloaded Image 1
Downloaded Image 2
Downloaded Image 4
Downloaded Image 0
Downloaded Image 3
End

操作系统的信号量是个很重要的概念，在进程控制方面都有应用。Java并发库的Semaphore可以很轻松完成信号量控制，Semaphore 可以控制某个资源可被同时访问的个数，acquire()获取一个许可，如果没有就等待，而release()释放一个许可。比如在Windows下可以 设置共享文件的最大客户端访问个数。

Semaphore维护了当前访问的个数，提供同步机制，控制同时访问的个数。在数据结构中链表可以保存“无限”的节点，用Semaphore可 以实现有限大小的链表。另外重入锁ReentrantLock也可以实现该功能，但实现上要负责些，代码也要复杂些。

下面的Demo中申明了一个只有5个许可的Semaphore，而有20个线程要访问这个资源，通过acquire()和release()获取 和释放访问许可。

 package  concurrent;

 import  java.util.concurrent.ExecutorService;
 import  java.util.concurrent.Executors;
 import  java.util.concurrent.Semaphore;

 public   class  TestSemaphore {
 public   static   void  main(String[] args) {
 // 线程池
 ExecutorService exec = Executors.newCachedThreadPool();
 // 只能5个线程同时访问
 final  Semaphore semp =  new  Semaphore( 5 );
 // 模拟20个客户端访问
 for  ( int  index =  0 ; index <  20 ; index++) {
 final   int  NO = index;
 Runnable run = new  Runnable() {
 public   void  run() {
 try  {
 // 获取许可
 semp.acquire();
 System.out.println(“Accessing: ” + NO);
 Thread.sleep((long ) (Math.random() *  10000 ));
 // 访问完后，释放
 semp.release();
 } catch  (InterruptedException e) {
 }
 }
 };
 exec.execute(run);
 }
 // 退出线程池
 exec.shutdown();
 }
 }

package concurrent;

import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Semaphore;

public class TestSemaphore {
public static void main(String[] args) {
// 线程池
ExecutorService exec = Executors.newCachedThreadPool();
// 只能5个线程同时访问
final Semaphore semp = new Semaphore(5);
// 模拟20个客户端访问
for (int index = 0; index < 20; index++) {
final int NO = index;
Runnable run = new Runnable() {
public void run() {
try {
// 获取许可
semp.acquire();
System.out.println(“Accessing: ” + NO);
Thread.sleep((long) (Math.random() * 10000));
// 访问完后，释放
semp.release();
} catch (InterruptedException e) {
}
}
};
exec.execute(run);
}
// 退出线程池
exec.shutdown();
}
}

运行结果：
Accessing: 0
Accessing: 1
Accessing: 2
Accessing: 3
Accessing: 4
Accessing: 5
Accessing: 6
Accessing: 7
Accessing: 8
Accessing: 9
Accessing: 10
Accessing: 11
Accessing: 12
Accessing: 13
Accessing: 14
Accessing: 15
Accessing: 16
Accessing: 17
Accessing: 18
Accessing: 19