worker模式_JAVA_编程开发_程序员俱乐部

中国优秀的程序员网站程序员频道CXYCLUB技术地图
热搜:
更多>>
 
您所在的位置: 程序员俱乐部 > 编程开发 > JAVA > worker模式

worker模式

 2018/3/29 22:01:22  huangjinjin520  程序员俱乐部  我要评论(0)
  • 摘要:今天来学学,大家也好对线程池有一个更好的理解。publicclassMain{publicstaticvoidmain(String[]args){Channelchannel=newChannel(5);//工人线程的數量,即线程池内的线程数目channel.startWorkers();//启动线程池内的线程newClientThread("Alice",channel).start();//发送请求的线程,相当于向队列加入请求newClientThread("Bobby",channel
  • 标签:模式
今天来学学,大家也好对线程池有一个更好的理解
public class Main { 
    public static void main(String[] args) { 
        Channel channel = new Channel(5);   // 工人线程的數量,即线程池内的线程数目  
        channel.startWorkers();//启动线程池内的线程  
        new ClientThread("Alice", channel).start();//发送请求的线程,相当于向队列加入请求  
        new ClientThread("Bobby", channel).start(); 
        new ClientThread("Chris", channel).start(); 
    } 


发送请求的client代码:
public class ClientThread extends Thread { 
    private final Channel channel;//相当于线程池  
     private static final Random random = new Random(); 
      public ClientThread(String name, Channel channel) { 
        super(name); 
        this.channel = channel; 
    } 
      public void run() { 
        try { 
            int i = 0; 
            Request request = new Request(getName(), i);//生成请求  
            channel.putRequest(request);//向队列中放入请求,也即把请求传给线程池  
            Thread.sleep(random.nextInt(1000)); 
        } catch (InterruptedException e) { 
        } 
    } 


ClientThread建立请求,并把请求传给了channel,下面来看看channel类(相当于线程池类)
public class Channel { 
    private static final int MAX_REQUEST = 100; 
    private final Request[] requestQueue;//存放请求的队列  
    private int tail;  // 下一个putRequest的地方 
    private int head;  // 下一个takeRequest的地方 
    private int count; // Request的数量  
      private final WorkerThread[] threadPool; 
      public Channel(int threads) { 
        this.requestQueue = new Request[MAX_REQUEST]; 
        this.head = 0; 
        this.tail = 0; 
        this.count = 0; 
          threadPool = new WorkerThread[threads]; 
        for (int i = 0; i < threadPool.length; i++) { 
            threadPool[i] = new WorkerThread("Worker-" + i, this);//生成线程池中的线程  
        } 
    } 
    public void startWorkers() { 
        for (int i = 0; i < threadPool.length; i++) { 
            threadPool[i].start();//启动线程池中的线程  
        } 
    } 
    public synchronized void putRequest(Request request) {//向队列中存入请求  
        while (count >= requestQueue.length) { 
            try { 
                wait(); 
            } catch (InterruptedException e) { 
            } 
        } 
        requestQueue[tail] = request; 
        tail = (tail + 1) % requestQueue.length; 
        count++; 
        notifyAll(); 
    } 
    public synchronized Request takeRequest() {//从队列取出请求  
        while (count <= 0) { 
            try { 
                wait(); 
            } catch (InterruptedException e) { 
            } 
        } 
        Request request = requestQueue[head]; 
        head = (head + 1) % requestQueue.length; 
        count--; 
        notifyAll(); 
        return request; 
    } 

channel类把传给他的请求放入队列中,等待worker去取请求,下面看看worker(即工作线程,线程池中已经初始话好的线程)
public class WorkerThread extends Thread { 
    private final Channel channel; 
    public WorkerThread(String name, Channel channel) { 
        super(name); 
        this.channel = channel; 
    } 
    public void run() { 
        while (true) { 
            Request request = channel.takeRequest();//取出请求  
            request.execute();//处理请求  
        } 
    } 

在工作线程中会从线程池的队列里取出请求,并对请求进行处理。这里的workerthread相当于背景线程,他一直都在运行,当有请求的时候,他就会进行处理,这里处理请求的线程是已经存在在channel(线程池里的线程),他不会因为请求的增加而增加(这是本例中的情况),不会来一个请求就新建立一个线程,节省了资源。

再看看请求的代码:
public class Request { 
    private final String name; //  委托者 
    private final int number;  // 请求编号  
    private static final Random random = new Random(); 
    public Request(String name, int number) { 
        this.name = name; 
        this.number = number; 
    } 
    public void execute() {//执行请求  
        System.out.println(Thread.currentThread().getName() + " executes " + this); 
        try { 
            Thread.sleep(random.nextInt(1000)); 
        } catch (InterruptedException e) { 
        } 
    } 
    public String toString() { 
        return "[ Request from " + name + " No." + number + " ]"; 
    } 

JAVA SDK所写的 ExecutorService,其就相当于channel,即线程池。至于其实现当然要比channel复杂多了,channel只是举个例子。而WorkerThread可不是工作线程,他相当于发送到channel的请求,也就是request,当执行代码:tpes.execute(workers[i]);时,相当于向线程池加入一个请求,而WorkerThread中的run则相当于request中的execute,这也是当执行tpes.execute(workers[i]);时,并不会产生新的线程的原因。ExecutorService中产生的背景线程(相当于本篇的WorkerThread )我们是看不到的。

各种it视频获取


  • 大小: 39.9 KB
  • 查看图片附件
发表评论
用户名: 匿名