Netty 工具类 —— HashedWheelTimer 讲解

一、前言

在网络通信中管理上万的连接，每个连接都有超时任务，如果为每个任务启动一个TImer超时器，那么会占用大量资源。为了解决这个问题，可用Netty工具类HashedWheelTimer。

二、HashedWheelTimer原理

1.概论

（学习一个类，最好的方式是看api文档或源码的注释，我下载了Netty源码）

这个类用来计划执行非精准的I/O超时。可以通过指定每一格的时间间隔来改变执行时间的精确度。在大多数网络应用中，I/O超时不需要十分准确，因此，默认的时间间隔是100 毫秒，这个值适用于大多数场合。HashedWheelTimer内部结构可以看做是个车轮，简单来说，就是TimerTask的hashTable的车轮。车轮的size默认是512，可以通过构造函数自己设置这个值。注意，当HashedWheelTimer被实例化启动后，会创建一个新的线程，因此，你的项目里应该只创建它的唯一一个实例。

（这个类的源自一位教授的论文 Tony Lauck's paper。算法是代码的灵魂，最难的是算法）

2.结构

下方图示阐述了大致的结构模型

Demo

@Test(timeout = 50000000)
    public void testTimerOverflowWheelLength() throws InterruptedException {
        final HashedWheelTimer timer = new HashedWheelTimer(
            Executors.defaultThreadFactory(), 100, TimeUnit.MILLISECONDS, 32);

        timer.newTimeout(new TimerTask() {
            @Override
            public void run(final Timeout timeout) throws Exception {
                System.out.println("lee");   //打印名字
            }
        }, 1000, TimeUnit.MILLISECONDS);

        Thread.sleep(10000);
        assertFalse(timer.stop().isEmpty());
    }

通过上面demo，我再说明一下hashedWheelTimer类的构造。

超时任务1000毫秒，超时之后，由hashedWheelTimer类中的worker线程，执行超时之后的任务（打印名字）。hashedWheelTimer有32个槽（相当于一个圆的32分之一），每移动一个槽的时间是100毫秒。
任务需要经过的tick数为: 1000 / 100 = 10次         (等待时长 / tickDuration)
任务需要经过的轮数为  : 10次 / 32次/轮 = 0轮     (tick总次数 / ticksPerWheel)
因为任务超时后不能马上被worker线程执行，需要等到worker线程移到相应卡槽位置时才会执行，因此说执行时间不精确。

hashedWheelTimer的核心是Worker线程，主要负责每过tickDuration时间就累加一次tick. 同时, 也负责执行到期的timeout任务, 此外，还负责添加timeou任务到指定的wheel中。

接下看看源码部分。

构造器

构造器的各个参数的说明和注释十分详细，应该没有不好理解的地方。

    /**
     * Creates a new timer.
     *
     * @param threadFactory        a {@link ThreadFactory} that creates a
     *                             background {@link Thread} which is dedicated to    用来创建后台线程
     *                             {@link TimerTask} execution.
     * @param tickDuration         the duration between tick                          每格时间间隔 默认 100
     * @param unit                 the time unit of the {@code tickDuration}          时间单位 默认 毫秒
     * @param ticksPerWheel        the size of the wheel                              轮子size（一圈多少格）
                                                                                         默认 512 如果不是2的N次方，则去大于该参数的第一个2的N次方
                                                                                         理由 便于Hash值的计算10      * @param leakDetection        {@code true} if leak detection should be enabled always,
     *                             if false it will only be enabled if the worker thread is not
     *                             a daemon thread.
     * @param  maxPendingTimeouts  The maximum number of pending timeouts after which call to
     *                             {@code newTimeout} will result in
     *                             {@link java.util.concurrent.RejectedExecutionException}
     *                             being thrown. No maximum pending timeouts limit is assumed if
     *                             this value is 0 or negative.                       最大待定超时时间 默认 不设置
     * @throws NullPointerException     if either of {@code threadFactory} and {@code unit} is {@code null}
     * @throws IllegalArgumentException if either of {@code tickDuration} and {@code ticksPerWheel} is <= 0
     */
    public HashedWheelTimer(
            ThreadFactory threadFactory,
            long tickDuration, TimeUnit unit, int ticksPerWheel, boolean leakDetection,
            long maxPendingTimeouts) {
           ...

        // Normalize ticksPerWheel to power of two and initialize the wheel.
        wheel = createWheel(ticksPerWheel);
        mask = wheel.length - 1;

        // Convert tickDuration to nanos.
        long duration = unit.toNanos(tickDuration);
           ...52 
        if (duration < MILLISECOND_NANOS) {
            if (logger.isWarnEnabled()) {
                logger.warn("Configured tickDuration %d smaller then %d, using 1ms.",
                            tickDuration, MILLISECOND_NANOS);
            }
            this.tickDuration = MILLISECOND_NANOS;
        } else {
            this.tickDuration = duration;
        }
        // 这里创建worker线程，worker线程是重点。
        workerThread = threadFactory.newThread(worker);

        leak = leakDetection || !workerThread.isDaemon() ? leakDetector.track(this) : null;

        this.maxPendingTimeouts = maxPendingTimeouts;

        if (INSTANCE_COUNTER.incrementAndGet() > INSTANCE_COUNT_LIMIT &&
            WARNED_TOO_MANY_INSTANCES.compareAndSet(false, true)) {
            reportTooManyInstances();
        }
    }

我们接下来看newTimeout（）方法

    @Override
    public Timeout newTimeout(TimerTask task, long delay, TimeUnit unit) {
           ... 9 
        long pendingTimeoutsCount = pendingTimeouts.incrementAndGet();

        if (maxPendingTimeouts > 0 && pendingTimeoutsCount > maxPendingTimeouts) {
            pendingTimeouts.decrementAndGet();
            throw new RejectedExecutionException("Number of pending timeouts ("
                + pendingTimeoutsCount + ") is greater than or equal to maximum allowed pending "
                + "timeouts (" + maxPendingTimeouts + ")");
        }

        start();

        // Add the timeout to the timeout queue which will be processed on the next tick.
        // During processing all the queued HashedWheelTimeouts will be added to the correct HashedWheelBucket.
        long deadline = System.nanoTime() + unit.toNanos(delay) - startTime;

        // Guard against overflow.
        if (delay > 0 && deadline < 0) {
            deadline = Long.MAX_VALUE;
        }
        HashedWheelTimeout timeout = new HashedWheelTimeout(this, task, deadline);
        timeouts.add(timeout);
        return timeout;
    }

我们接着进到worker线程的启动：

/**
     * Starts the background thread explicitly.  The background thread will
     * start automatically on demand even if you did not call this method.
     *
     * @throws IllegalStateException if this timer has been
     *                               {@linkplain #stop() stopped} already
     */
    public void start() {     // 这个方法为什么是public的？我们可以在实例化HashedWheelTimer后，主动调用这个方法，启动worker线程
        switch (WORKER_STATE_UPDATER.get(this)) {
            case WORKER_STATE_INIT:
                if (WORKER_STATE_UPDATER.compareAndSet(this, WORKER_STATE_INIT, WORKER_STATE_STARTED)) {
                    workerThread.start();            // 在这里启动了worker线程
                }
                break;
            case WORKER_STATE_STARTED:
                break;
            case WORKER_STATE_SHUTDOWN:
                throw new IllegalStateException("cannot be started once stopped");
            default:
                throw new Error("Invalid WorkerState");
        }

        // Wait until the startTime is initialized by the worker.
           // 为了等待worker线程初始化startTime
        while (startTime == 0) {
            try {
                startTimeInitialized.await();
            } catch (InterruptedException ignore) {
                // Ignore - it will be ready very soon.
            }
        }
    }

 下一步，我们看看worker线程里面的操作。

首先是初始化startTime，CountDownLatch的触发，后面do while操作可以看作是圆盘在一个个转动，每转一个就会用worker线程处理，格子中超时的任务。

        @Override
        public void run() {
            // Initialize the startTime.
            startTime = System.nanoTime();
            if (startTime == 0) {
                // We use 0 as an indicator for the uninitialized value here, so make sure it's not 0 when initialized.
                startTime = 1;
            }

            // Notify the other threads waiting for the initialization at start().
            startTimeInitialized.countDown();

            do {
                final long deadline = waitForNextTick();
                if (deadline > 0) {
                    int idx = (int) (tick & mask);
                    processCancelledTasks();
                    HashedWheelBucket bucket =
                            wheel[idx];
                    transferTimeoutsToBuckets();
                    bucket.expireTimeouts(deadline);
                    tick++;
                }
            } while (WORKER_STATE_UPDATER.get(HashedWheelTimer.this) == WORKER_STATE_STARTED);

            // Fill the unprocessedTimeouts so we can return them from stop() method.
            for (HashedWheelBucket bucket: wheel) {
                bucket.clearTimeouts(unprocessedTimeouts);
            }
            for (;;) {
                HashedWheelTimeout timeout = timeouts.poll();
                if (timeout == null) {
                    break;
                }
                if (!timeout.isCancelled()) {
                    unprocessedTimeouts.add(timeout);
                }
            }
            processCancelledTasks();
        }

----------------------------------------------------------------------------------------------

 参考资源：

https://www.jianshu.com/p/328f22432638

https://my.oschina.net/haogrgr/blog/489320

https://chuansongme.com/n/1650380646616