**
**
生产者和消费者问题是线程模型中的经典问题:生产者和消费者在同一时间段内共用同一个存储空间,生产者往存储空间中添加产品,消费者从存储空间中取走产品,当存储空间为空时,消费者阻塞,当存储空间满时,生产者阻塞。
以下这些解法,其实本质上都是实现了一个阻塞队列。为空,则消费者阻塞,满了,则生产者阻塞。
**
**
这也是最简单最基础的实现,缓冲区满和为空时都调用wait()方法等待,当生产者生产了一个产品或者消费者消费了一个产品之后会唤醒所有线程。
public static void testProductConsumeByWaitAndNotify() {
final int size = 10;
final Queue<String> queue = new ArrayDeque<String>(size);
final Object lock = new Object();
Runnable producer = new Runnable() {
public void run() {
for(int i=0;i<30;i++) {
try {
Thread.sleep(100);
} catch (InterruptedException e) {
e.printStackTrace();
}
String msg = "消息:"+i;
//队列未满,一直往里放消息
synchronized (lock) {
while (size == queue.size()) {
try {
lock.wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
queue.offer(msg);
lock.notifyAll();
}
System.out.println(msg+" 已发送");
}
}
};
Runnable consumer = new Runnable() {
public void run() {
while (true) {
try {
Thread.sleep(200);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
synchronized (lock) {
while (queue.size() == 0) {
try {
lock.wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
String msg = queue.poll();
System.out.println(msg+"已消费");
lock.notifyAll();
}
}
}
};
new Thread(producer).start();
new Thread(producer).start();
new Thread(producer).start();
new Thread(consumer).start();
new Thread(consumer).start();
}
**
**
java.util.concurrent.lock 中的 Lock 框架是锁定的一个抽象,通过对lock的lock()方法和unlock()方法实现了对锁的显示控制,而synchronize()则是对锁的隐性控制。
可重入锁,也叫做递归锁,指的是同一线程 外层函数获得锁之后 ,内层递归函数仍然有获取该锁的代码,但不受影响,简单来说,该锁维护这一个与获取锁相关的计数器,如果拥有锁的某个线程再次得到锁,那么获取计数器就加1,函数调用结束计数器就减1,然后锁需要被释放两次才能获得真正释放。已经获取锁的线程进入其他需要相同锁的同步代码块不会被阻塞。
ReentrantLock的Condition:
//阻塞当前线程,直到收到通知或者被中断(将当前线程加入到当前Condition对象的等待队列里)
//Block until signalled or interrupted
public final void await() throws InterruptedException;
/**
* Moves the longest-waiting thread, if one exists, from the
* wait queue for this condition to the wait queue for the
* owning lock.
* 把在当前Condition对象的等待队列里的等待最久的线程,转移到当前Lock的等待队列里
* @throws IllegalMonitorStateException if {@link #isHeldExclusively}
* returns {@code false}
*/
public final void signal() ;
ReentrantLock实现生产消费模型:
public static void testProductConsumeByLock() {
final Lock lock = new ReentrantLock();
final Condition empty = lock.newCondition();
final Condition full = lock.newCondition();
final int size = 10;
final Queue<String> queue = new ArrayDeque<String>(size);
Runnable producer = new Runnable() {
public void run() {
try {
Thread.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
for(int i=0;i<20;i++) {
lock.lock();
try {
if(queue.size() == size) {
try {
full.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
String msg = "生产消息:"+i;
queue.add(msg);
System.out.println(msg);
empty.signal();
} finally {
lock.unlock();
}
}
}
};
Runnable consumer = new Runnable() {
public void run() {
try {
Thread.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
while (true) {
lock.lock();
try {
if(queue.isEmpty()) {
try {
empty.await();
} catch (InterruptedException e) {
e.printStackTrace();
}
}else {
String msg = queue.remove();
System.out.println(msg + "已消费");
full.signal();
}
} finally {
lock.unlock();
}
}
}
};
new Thread(producer).start();
new Thread(producer).start();
new Thread(producer).start();
new Thread(consumer).start();
new Thread(consumer).start();
}
**
**
BlockingQueue即阻塞队列,从阻塞这个词可以看出,在某些情况下对阻塞队列的访问可能会造成阻塞。被阻塞的情况主要有如下两种:
因此,当一个线程对已经满了的阻塞队列进行入队操作时会阻塞,除非有另外一个线程进行了出队操作,当一个线程对一个空的阻塞队列进行出队操作时也会阻塞,除非有另外一个线程进行了入队操作。
从上可知,阻塞队列是线程安全的。
下面是BlockingQueue接口的一些方法:
操作 | 抛异常 | 特定值 | 阻塞 | 超时 |
---|---|---|---|---|
插入 | add(o) | offer(o) | put(o) | offer(o, timeout, timeunit) |
移除 | remove(o) | poll(o) | take(o) | poll(timeout, timeunit) |
检查 | element(o) | peek(o) |
这四类方法分别对应的是:
下面来看由阻塞队列实现的生产消费模型,这里我们使用take()和put()方法,这里生产者和生产者,消费者和消费者之间不存在同步,所以会出现连续生成和连续消费的现象
/**
* 生产者消费者
* 使用阻塞队列实现
* @throws InterruptedException
*/
public static void testProductConsumeByBlockingQueue() throws InterruptedException {
//因为SynchronousQueue没有存储功能,因此put和take会一直阻塞,直到有另一个线程已经准备好参与到交付过程中。仅当有足够多的消费者,并且总是有一个消费者准备好获取交付的工作时,才适合使用同步队列。
// final BlockingQueue queue = new SynchronousQueue(true);
//使用有界阻塞队列
final BlockingQueue<String> queue = new ArrayBlockingQueue<String>(10);
Runnable producer = new Runnable() {
public void run() {
for(int i=0;i<100;i++) {
try {
Thread.sleep(100);
} catch (InterruptedException e) {
e.printStackTrace();
}
String msg = "消息:"+i;
try {
queue.put(msg);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
System.out.println(msg+" 已发送");
}
}
};
Runnable consumer = new Runnable() {
public void run() {
while (true) {
try {
Thread.sleep(200);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
String msg = null;
try {
msg = queue.take();
} catch (InterruptedException e) {
e.printStackTrace();
}
System.out.println(msg+"已消费");
}
}
};
new Thread(producer).start();
new Thread(consumer).start();
}
**
**
信号量可以控制访问相应资源的线程的数量,从而实现生产消费模型
import java.util.concurrent.Semaphore;
public class BySemaphore {
int count = 0;
final Semaphore put = new Semaphore(5);// 初始令牌个数
final Semaphore get = new Semaphore(0);
final Semaphore mutex = new Semaphore(1); //该信号量相当于锁
public static void main(String[] args) {
BySemaphore bySemaphore = new BySemaphore();
new Thread(bySemaphore.new Producer()).start();
new Thread(bySemaphore.new Consumer()).start();
new Thread(bySemaphore.new Consumer()).start();
new Thread(bySemaphore.new Producer()).start();
}
class Producer implements Runnable {
@Override
public void run() {
for (int i = 0; i < 5; i++) {
try {
Thread.sleep(1000);
} catch (Exception e) {
e.printStackTrace();
}
try {
put.acquire();// 注意顺序
mutex.acquire();
count++;
System.out.println("生产者" + Thread.currentThread().getName()
+ "已生产完成,商品数量:" + count);
} catch (Exception e) {
e.printStackTrace();
} finally {
mutex.release();
get.release();
}
}
}
}
class Consumer implements Runnable {
@Override
public void run() {
for (int i = 0; i < 5; i++) {
try {
Thread.sleep(1000);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
try {
get.acquire();// 注意顺序
mutex.acquire();
count--;
System.out.println("消费者" + Thread.currentThread().getName()
+ "已消费,剩余商品数量:" + count);
} catch (Exception e) {
e.printStackTrace();
} finally {
mutex.release();
put.release();
}
}
}
}
}
**
**
这个是取巧的办法,直接使用现成的消息中间件服务(如RocketMq、RabbitMq、Kafka等),分分钟搞定。手动微笑~~