Familiarity with Java multi-threaded programming students know that when we create a thread too easily lead to memory overflow, so we will need to use the thread pool technology. I read some recent related articles, and personally studied at the source and found some articles still some problems, so I summed up, in this dedicated to you.
1 thread pool advantage
Overall, the thread pool has the following advantages: (1) reduce resource consumption . By reusing the thread has been created to reduce thread creation and destruction caused by consumption. (2) increase the response speed . When the mission arrives, the task may not need to wait until the thread creation can be implemented immediately. (3) increase the thread manageability . A thread is a scarce resource, if the unlimited creation, not only consumes system resources, but also reduce the stability of the system, using a thread pool can be unified distribution, tuning and monitoring.
2 Use the thread pool
Real thread pool class is ThreadPoolExecutor, configured with the following four kinds of methods:
public ThreadPoolExecutor(int corePoolSize,
int maximumPoolSize,
long keepAliveTime,
TimeUnit unit,
BlockingQueue<Runnable> workQueue) {
this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue,
Executors.defaultThreadFactory(), defaultHandler);
}
public ThreadPoolExecutor(int corePoolSize,
int maximumPoolSize,
long keepAliveTime,
TimeUnit unit,
BlockingQueue<Runnable> workQueue,
ThreadFactory threadFactory) {
this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue,
threadFactory, defaultHandler);
}
public ThreadPoolExecutor(int corePoolSize,
int maximumPoolSize,
long keepAliveTime,
TimeUnit unit,
BlockingQueue<Runnable> workQueue,
RejectedExecutionHandler handler) {
this(corePoolSize, maximumPoolSize, keepAliveTime, unit, workQueue,
Executors.defaultThreadFactory(), handler);
}
public ThreadPoolExecutor(int corePoolSize,
int maximumPoolSize,
long keepAliveTime,
TimeUnit unit,
BlockingQueue<Runnable> 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;
}
可以看到，其需要如下几个参数：
- corePoolSize（必需）：核心线程数。默认情况下，核心线程会一直存活，但是当将allowCoreThreadTimeout设置为true时，核心线程也会超时回收。
- maximumPoolSize（必需）：线程池所能容纳的最大线程数。当活跃线程数达到该数值后，后续的新任务将会阻塞。
- keepAliveTime（必需）：线程闲置超时时长。如果超过该时长，非核心线程就会被回收。如果将allowCoreThreadTimeout设置为true时，核心线程也会超时回收。
- unit（必需）：指定keepAliveTime参数的时间单位。常用的有：TimeUnit.MILLISECONDS（毫秒）、TimeUnit.SECONDS（秒）、TimeUnit.MINUTES（分）。
- workQueue（必需）：任务队列。通过线程池的execute()方法提交的Runnable对象将存储在该参数中。
- threadFactory（可选）：线程工厂。用于指定为线程池创建新线程的方式。
- handler（可选）：拒绝策略。当达到最大线程数时需要执行的饱和策略。
线程池的使用流程如下：
// 创建线程池
Executor threadPool = new ThreadPoolExecutor(CORE_POOL_SIZE,
MAXIMUM_POOL_SIZE,
KEEP_ALIVE,
TimeUnit.SECONDS,
sPoolWorkQueue,
sThreadFactory);
// 向线程池提交任务
threadPool.execute(new Runnable() {
@Override
public void run() {
... // 线程执行的任务
}
});
// 关闭线程池
threadPool.shutdown(); // 设置线程池的状态为SHUTDOWN，然后中断所有没有正在执行任务的线程
threadPool.shutdownNow(); // 设置线程池的状态为 STOP，然后尝试停止所有的正在执行或暂停任务的线程，并返回等待执行任务的列表
3 线程池的工作原理
下面来描述一下线程池工作的原理，同时对上面的参数有一个更深的了解。其工作原理流程图如下：
通过上图，相信大家已经对所有参数有个了解了。其实还有一点，在线程池中并没有区分线程是否是核心线程的。下面我们再对任务队列、线程工厂和拒绝策略做更多的说明。
4 线程池的参数
4.1 任务队列（workQueue）
任务队列是基于阻塞队列实现的，即采用生产者消费者模式，在Java中需要实现BlockingQueue接口。但Java已经为我们提供了7种阻塞队列的实现：
- ArrayBlockingQueue：一个由数组结构组成的有界阻塞队列（数组结构可配合指针实现一个环形队列）。
- LinkedBlockingQueue： 一个由链表结构组成的有界阻塞队列，在未指明容量时，容量默认为Integer.MAX_VALUE。
- PriorityBlockingQueue： 一个支持优先级排序的无界阻塞队列，对元素没有要求，可以实现Comparable接口也可以提供Comparator来对队列中的元素进行比较。跟时间没有任何关系，仅仅是按照优先级取任务。
- DelayQueue：类似于PriorityBlockingQueue，是二叉堆实现的无界优先级阻塞队列。要求元素都实现Delayed接口，通过执行时延从队列中提取任务，时间没到任务取不出来。
- SynchronousQueue： 一个不存储元素的阻塞队列，消费者线程调用take()方法的时候就会发生阻塞，直到有一个生产者线程生产了一个元素，消费者线程就可以拿到这个元素并返回；生产者线程调用put()方法的时候也会发生阻塞，直到有一个消费者线程消费了一个元素，生产者才会返回。
- LinkedBlockingDeque： 使用双向队列实现的有界双端阻塞队列。双端意味着可以像普通队列一样FIFO（先进先出），也可以像栈一样FILO（先进后出）。
- LinkedTransferQueue： 它是ConcurrentLinkedQueue、LinkedBlockingQueue和SynchronousQueue的结合体，但是把它用在ThreadPoolExecutor中，和LinkedBlockingQueue行为一致，但是是无界的阻塞队列。
注意有界队列和无界队列的区别：如果使用有界队列，当队列饱和时并超过最大线程数时就会执行拒绝策略；而如果使用无界队列，因为任务队列永远都可以添加任务，所以设置maximumPoolSize没有任何意义。
4.2 线程工厂（threadFactory）
线程工厂指定创建线程的方式，需要实现ThreadFactory接口，并实现newThread(Runnable r)方法。该参数可以不用指定，Executors框架已经为我们实现了一个默认的线程工厂：
/**
* The default thread factory.
*/
private static class DefaultThreadFactory implements ThreadFactory {
private static final AtomicInteger poolNumber = new AtomicInteger(1);
private final ThreadGroup group;
private final AtomicInteger threadNumber = new AtomicInteger(1);
private final String namePrefix;
DefaultThreadFactory() {
SecurityManager s = System.getSecurityManager();
group = (s != null) ? s.getThreadGroup() :
Thread.currentThread().getThreadGroup();
namePrefix = "pool-" +
poolNumber.getAndIncrement() +
"-thread-";
}
public Thread newThread(Runnable r) {
Thread t = new Thread(group, r,
namePrefix + threadNumber.getAndIncrement(),
0);
if (t.isDaemon())
t.setDaemon(false);
if (t.getPriority() != Thread.NORM_PRIORITY)
t.setPriority(Thread.NORM_PRIORITY);
return t;
}
}
4.3 拒绝策略（handler）
当线程池的线程数达到最大线程数时，需要执行拒绝策略。拒绝策略需要实现RejectedExecutionHandler接口，并实现rejectedExecution(Runnable r, ThreadPoolExecutor executor)方法。不过Executors框架已经为我们实现了4种拒绝策略：
- AbortPolicy（默认）：丢弃任务并抛出RejectedExecutionException异常。
- CallerRunsPolicy：由调用线程处理该任务。
- DiscardPolicy：丢弃任务，但是不抛出异常。可以配合这种模式进行自定义的处理方式。
- DiscardOldestPolicy：丢弃队列最早的未处理任务，然后重新尝试执行任务。
5 功能线程池
嫌上面使用线程池的方法太麻烦？其实Executors已经为我们封装好了4种常见的功能线程池，如下：
- 定长线程池（FixedThreadPool）
- 定时线程池（ScheduledThreadPool ）
- 可缓存线程池（CachedThreadPool）
- 单线程化线程池（SingleThreadExecutor）
5.1 定长线程池（FixedThreadPool）
创建方法的源码：
public static ExecutorService newFixedThreadPool(int nThreads) {
return new ThreadPoolExecutor(nThreads, nThreads,
0L, TimeUnit.MILLISECONDS,
new LinkedBlockingQueue<Runnable>());
}
public static ExecutorService newFixedThreadPool(int nThreads, ThreadFactory threadFactory) {
return new ThreadPoolExecutor(nThreads, nThreads,
0L, TimeUnit.MILLISECONDS,
new LinkedBlockingQueue<Runnable>(),
threadFactory);
}
- 特点：只有核心线程，线程数量固定，执行完立即回收，任务队列为链表结构的有界队列。
- 应用场景：控制线程最大并发数。
使用示例：
// 1. 创建定长线程池对象 & 设置线程池线程数量固定为3
ExecutorService fixedThreadPool = Executors.newFixedThreadPool(3);
// 2. 创建好Runnable类线程对象 & 需执行的任务
Runnable task =new Runnable(){
public void run() {
System.out.println("执行任务啦");
}
};
// 3. 向线程池提交任务
fixedThreadPool.execute(task);
5.2 定时线程池（ScheduledThreadPool ）
创建方法的源码：
private static final long DEFAULT_KEEPALIVE_MILLIS = 10L;
public static ScheduledExecutorService newScheduledThreadPool(int corePoolSize) {
return new ScheduledThreadPoolExecutor(corePoolSize);
}
public ScheduledThreadPoolExecutor(int corePoolSize) {
super(corePoolSize, Integer.MAX_VALUE,
DEFAULT_KEEPALIVE_MILLIS, MILLISECONDS,
new DelayedWorkQueue());
}
public static ScheduledExecutorService newScheduledThreadPool(
int corePoolSize, ThreadFactory threadFactory) {
return new ScheduledThreadPoolExecutor(corePoolSize, threadFactory);
}
public ScheduledThreadPoolExecutor(int corePoolSize,
ThreadFactory threadFactory) {
super(corePoolSize, Integer.MAX_VALUE,
DEFAULT_KEEPALIVE_MILLIS, MILLISECONDS,
new DelayedWorkQueue(), threadFactory);
}
- 特点：核心线程数量固定，非核心线程数量无限，执行完闲置10ms后回收，任务队列为延时阻塞队列。
- 应用场景：执行定时或周期性的任务。
使用示例：
// 1. 创建 定时线程池对象 & 设置线程池线程数量固定为5
ScheduledExecutorService scheduledThreadPool = Executors.newScheduledThreadPool(5);
// 2. 创建好Runnable类线程对象 & 需执行的任务
Runnable task =new Runnable(){
public void run() {
System.out.println("执行任务啦");
}
};
// 3. 向线程池提交任务
scheduledThreadPool.schedule(task, 1, TimeUnit.SECONDS); // 延迟1s后执行任务
scheduledThreadPool.scheduleAtFixedRate(task,10,1000,TimeUnit.MILLISECONDS);// 延迟10ms后、每隔1000ms执行任务
5.3 可缓存线程池（CachedThreadPool）
创建方法的源码：
public static ExecutorService newCachedThreadPool() {
return new ThreadPoolExecutor(0, Integer.MAX_VALUE,
60L, TimeUnit.SECONDS,
new SynchronousQueue<Runnable>());
}
public static ExecutorService newCachedThreadPool(ThreadFactory threadFactory) {
return new ThreadPoolExecutor(0, Integer.MAX_VALUE,
60L, TimeUnit.SECONDS,
new SynchronousQueue<Runnable>(),
threadFactory);
}
- 特点：无核心线程，非核心线程数量无限，执行完闲置60s后回收，任务队列为不存储元素的阻塞队列。
- 应用场景：执行大量、耗时少的任务。
使用示例：
// 1. 创建可缓存线程池对象
ExecutorService cachedThreadPool = Executors.newCachedThreadPool();
// 2. 创建好Runnable类线程对象 & 需执行的任务
Runnable task =new Runnable(){
public void run() {
System.out.println("执行任务啦");
}
};
// 3. 向线程池提交任务
cachedThreadPool.execute(task);
5.4 单线程化线程池（SingleThreadExecutor）
创建方法的源码：
public static ExecutorService newSingleThreadExecutor() {
return new FinalizableDelegatedExecutorService
(new ThreadPoolExecutor(1, 1,
0L, TimeUnit.MILLISECONDS,
new LinkedBlockingQueue<Runnable>()));
}
public static ExecutorService newSingleThreadExecutor(ThreadFactory threadFactory) {
return new FinalizableDelegatedExecutorService
(new ThreadPoolExecutor(1, 1,
0L, TimeUnit.MILLISECONDS,
new LinkedBlockingQueue<Runnable>(),
threadFactory));
}
- 特点：只有1个核心线程，无非核心线程，执行完立即回收，任务队列为链表结构的有界队列。
- 应用场景：不适合并发但可能引起IO阻塞性及影响UI线程响应的操作，如数据库操作、文件操作等。
使用示例：
// 1. Create a single-threaded thread pool
ExecutorService singleThreadExecutor = Executors.newSingleThreadExecutor ();
// 2. Runnable class created thread object & task you need to perform
Runnable Task = new new Runnable () {
public void RUN () {
System. out.println ( "mission friends");
}
};
// 3. submit jobs to the thread pool
singleThreadExecutor.execute (task);
5.5 Comparison
Summary 6
Executors of the four functions of the thread pool is convenient, but now is not recommended, but rather recommended directly by using ThreadPoolExecutor way, this approach allows the students to write more explicit operating rules thread pool, to avoid the risk of resource depletion .
In fact, Executors of four functional threads have the following drawbacks:
FixedThreadPool and SingleThreadExecutor: The main problem is the accumulation of request processing queues are used LinkedBlockingQueue, may consume very large memory, even OOM.
CachedThreadPool and ScheduledThreadPool: The main problem is the maximum number of threads is Integer.MAX_VALUE, may create a very large number of threads, even OOM.