• 谷粒商城 高级篇 (十三) --------- 异步&线程池



    一、线程回顾

    1. 初始化线程的 4 种方式

    A、继承 Thread

    /* 
    * 1)、继承Thread
    *     Thread01 thread = new Thread01();
    *     thread.start();//启动线程
    */
    public static class Thread01 extends Thread{
        @Override
        public void run() {
            System.out.println("当前线程:"+Thread.currentThread().getId());
            int i = 10 / 2;
            System.out.println("运行结果:"+i);
        }
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13

    B、实现 Runnable 接口

    /*
    * 2)、实现Runnable接口
    *     Runable01 runable01 = new Runable01();
    *     new Thread(runable01).start();
    */
    public static class Runable01 implements Runnable{
    
        @Override
        public void run() {
            System.out.println("当前线程:"+Thread.currentThread().getId());
            int i = 10 / 2;
            System.out.println("运行结果:"+i);
        }
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14

    C、实现 Callable 接口 + FutureTask (可以拿到返回结果,可以处理异常)

    /*
    * 3)、实现Callable接口 + FutureTask (可以拿到返回结果,可以处理异常)
    *      FutureTask futureTask = new FutureTask<>(new Callable01());
    *      new Thread(futureTask).start();
    *      //阻塞等待整个线程执行完成,获取返回结果
    *      Integer integer = futureTask.get();
    */
    public static class Callable01 implements Callable<Integer>{
        @Override
        public Integer call() throws Exception {
            System.out.println("当前线程:"+Thread.currentThread().getId());
            int i = 10 / 2;
            System.out.println("运行结果:"+i);
            return i;
        }
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16

    D、线程池

    方式 1 和方式 2:主进程无法获取线程的运算结果。不适合当前场景

    方式 3:主进程可以获取线程的运算结果,但是不利于控制服务器中的线程资源。可以导致服务器资源耗尽。

    方式 4:通过如下两种方式初始化线程池

    Executors.newFiexedThreadPool(3);
    //或者
    new ThreadPoolExecutor(corePoolSize, maximumPoolSize, keepAliveTime, TimeUnit unit, workQueue, threadFactory, handler);
    
    • 1
    • 2
    • 3

    通过线程池性能稳定,也可以获取执行结果,并捕获异常。但是,在业务复杂情况下,一个异步调用可能会依赖于另一个异步调用的执行结果。

    2. 线程池的七大参数

    * @param corePoolSize the number of threads to keep in the pool, even* if they are idle, unless {@code allowCoreThreadTimeOut} isset 池中一直保持的线程的数量,即使线程空闲。除非设置了 allowCoreThreadTimeOut
    
    * @param maximumPoolSize the maximum number of threads to allow inthe* pool 池中允许的最大的线程数
    
    * @param keepAliveTime when the number of threads is greater than* the core, this is the maximumtime that excess idle threads* will wait for new tasks before terminating. 当线程数大于核心线程数的时候,线程在最大多长时间没有接到新任务就会终止释放,最终线程池维持在 corePoolSize 大小
    
    * @param unit the time unit for the {@code keepAliveTime} argument时间单位
    
    * @param workQueue the queue to use for holding tasks before theyare* executed. This queue will hold only the {@code Runnable}* tasks submitted by the {@code execute} method. 阻塞队列,用来存储等待执行的任务,如果当前对线程的需求超过了corePoolSize大小,就会放在这里等待空闲线程执行。
    
    * @param threadFactory the factory to use when the executor creates a new thread创建线程的工厂,比如指定线程名等
    
    * @param handler the handler to use when execution is blocked because the thread bounds and queue capacities are reached拒绝策略,如果线程满了,线程池就会使用拒绝策略。
    
    
    //我们以后再业务代码里面,以上三种启动线程的方式都不用。【将所有的多线程异步任务都交给线程池执行】
    
    /*
    4)、线程池[ExecutorService]
    *     给线程池直接提交任务。
    *     service.execute(new Runable01());
    *     1、创建:
    *        1)、Executors
    *        2)、new ThreadPoolExecutor
             Future:可以获取到异步结果
    *
    * 区别:
    *     1、2不能得到返回值。3可以获取返回值
    *     1、2、3都不能控制资源
    *     4可以控制资源,性能稳定。
    */
    
    
     //当前系统中池只有一两个,每个异步任务,提交给线程池让他自己去执行就行
    /**
     * 七大参数
     * corePoolSize:[5] 核心线程数[一直存在除非(allowCoreThreadTimeOut)];  线程池,创建好以后就准备就绪的线程数量,就等待来接受异步任务去执行。
     * 		5个  Thread thread = new Thread();  thread.start();
     * maximumPoolSize:[200] 最大线程数量;  控制资源
     * keepAliveTime: 存活时间。如果当前的线程数量大于 core 数量。
     * 		释放空闲的线程(maximumPoolSize-corePoolSize)。只要线程空闲大于指定的keepAliveTime;
     * unit:时间单位
     * BlockingQueue workQueue:阻塞队列。如果任务有很多,就会将目前多的任务放在队列里面。只要有线程空闲,就会去队列里面取出新的任务继续执行。
     * threadFactory:线程的创建工厂。
     * RejectedExecutionHandler handler:如果队列满了,按照我们指定的拒绝策略拒绝执行任务
     */
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25
    • 26
    • 27
    • 28
    • 29
    • 30
    • 31
    • 32
    • 33
    • 34
    • 35
    • 36
    • 37
    • 38
    • 39
    • 40
    • 41
    • 42
    • 43
    • 44
    • 45
    • 46

    运行流程:

    1、线程池创建,准备好 core 数量的核心线程,准备接受任务

    2、新的任务进来,用 core 准备好的空闲线程执行。

    (1) 、core 满了,就将再进来的任务放入阻塞队列中。空闲的core 就会自己去阻塞队列获取任务执行
    (2) 、阻塞队列满了,就直接开新线程执行,最大只能开到 max 指定的数量
    (3) 、max 都执行好了。Max-core 数量空闲的线程会在 keepAliveTime 指定的时间后自动销毁。最终保持到 core 大小
    (4) 、如果线程数开到了 max 的数量,还有新任务进来,就会使用reject 指定的拒绝策略进行处理

    3、所有的线程创建都是由指定的 factory 创建的。

    public void thread(String[] args) throws ExecutionException, InterruptedException {
        System.out.println("main....start....");
        /**
         * 
       
    //    new Thread(()-> System.out.println("hello")).start();
    
       
         * 工作顺序:
         * 1)、线程池创建,准备好core数量的核心线程,准备接受任务
         * 1.1、core满了,就将再进来的任务放入阻塞队列中。空闲的core就会自己去阻塞队列获取任务执行
         * 1.2、阻塞队列满了,就直接开新线程执行,最大只能开到max指定的数量
         * 1.3、max满了就用RejectedExecutionHandler拒绝任务
         * 1.4、max都执行完成,有很多空闲.在指定的时间keepAliveTime以后,释放max-core这些线程
         *
         *      new LinkedBlockingDeque<>():默认是Integer的最大值。内存不够
         *
         * 一个线程池 core 7; max 20 ,queue:50,100并发进来怎么分配的;
         * 7个会立即得到执行,50个会进入队列,再开13个进行执行。剩下的30个就使用拒绝策略。
         * 如果不想抛弃还要执行。CallerRunsPolicy;
         *
         */
        ThreadPoolExecutor executor = new ThreadPoolExecutor(5,
                200,
                10,
                TimeUnit.SECONDS,
                new LinkedBlockingDeque<>(100000),
                Executors.defaultThreadFactory(),
                new ThreadPoolExecutor.AbortPolicy());
        System.out.println("main....end....");
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25
    • 26
    • 27
    • 28
    • 29
    • 30
    • 31

    面试:

    一个线程池 core 7; max 20 ,queue:50,100 并发进来怎么分配的;先有 7 个能直接得到执行,接下来 50
    个进入队列排队,在多开 13 个继续执行。现在70 个被安排上了。剩下 30 个默认拒绝策略。

    3. 常见的 4 种线程池

    A、newCachedThreadPool

    创建一个可缓存线程池,如果线程池长度超过处理需要,可灵活回收空闲线程,若无可回收,则新建线程。

    B、newFixedThreadPool

    创建一个定长线程池,可控制线程最大并发数,超出的线程会在队列中等待。

    C、newScheduledThreadPool

    创建一个定长线程池,支持定时及周期性任务执行。

    D、newSingleThreadExecutor

    创建一个单线程化的线程池,它只会用唯一的工作线程来执行任务,保证所有任务按照指定顺序 (FIFO、LIFO、优先级) 执行。

    //  Executors.newCachedThreadPool() core是0,所有都可回收
    //  Executors.newFixedThreadPool() 固定大小,core=max;都不可回收
    //  Executors.newScheduledThreadPool() 定时任务的线程池
    //  Executors.newSingleThreadExecutor() 单线程的线程池,后台从队列里面获取任务,挨个执行
    
    • 1
    • 2
    • 3
    • 4

    4. 开发中为什么使用线程池

    • 降低资源的消耗
    • 通过重复利用已经创建好的线程降低线程的创建和销毁带来的损耗
    • 提高响应速度
    • 因为线程池中的线程数没有超过线程池的最大上限时,有的线程处于等待分配任务的状态,当任务来时无需创建新的线程就能执行
    • 提高线程的可管理性
    • 线程池会根据当前系统特点对池内的线程进行优化处理,减少创建和销毁线程带来的系统开销。无限的创建和销毁线程不仅消耗系统资源,还降低系统的稳定性,使用线程池进行统一分配

    二、CompletableFuture 异步编排

    1. 业务场景

    查询商品详情页的逻辑比较复杂,有些数据还需要远程调用,必然需要花费更多的时间。假如商品详情页的每个查询,需要如下标注的时间才能完成。

    在这里插入图片描述

    那么,用户需要 5.5s 后才能看到商品详情页的内容。很显然是不能接受的。如果有多个线程同时完成这 6 步操作,也许只需要 1.5s 即可完成响应。Future 是 Java 5 添加的类,用来描述一个异步计算的结果。你可以使用isDone方法检查计 算是否完成,或者使用get阻塞住调用线程,直到计算完成返回结果,你也可以使用cancel 方法停止任务的执行。

    虽然Future以及相关使用方法提供了异步执行任务的能力,但是对于结果的获取却是很不方便,只能通过阻塞或者轮询的方式得到任务的结果。阻塞的方式显然和我们的异步编程的初衷相违背,轮询的方式又会耗费无谓的 CPU 资源,而且也不能及时地得到计算结果,为什么不能用观察者设计模式当计算结果完成及时通知监听者呢?

    很多语言,比如 Node.js,采用回调的方式实现异步编程。Java 的一些框架,比如Netty,自己扩展了 Java 的 Future接口,提供了addListener等多个扩展方法;Google guava 也提供了通用的扩展 Future;Scala 也提供了简单易用且功能强大的 Future/Promise 异步编程模式。

    作为正统的 Java 类库,是不是应该做点什么,加强一下自身库的功能呢?

    在 Java 8 中, 新增加了一个包含 50 个方法左右的类: CompletableFuture,提供了非常强大的Future 的扩展功能,可以帮助我们简化异步编程的复杂性,提供了函数式编程的能力,可以通过回调的方式处理计算结果,并且提供了转换和组合 CompletableFuture 的方法。CompletableFuture 类实现了 Future 接口,所以你还是可以像以前一样通过get方法阻塞或者轮询的方式获得结果,但是这种方式不推荐使用。

    CompletableFuture 和 FutureTask 同属于 Future 接口的实现类,都可以获取线程的执行结果。

    在这里插入图片描述

    2. 创建异步对象

    CompletableFuture 提供了四个静态方法来创建一个异步操作。

    在这里插入图片描述

    A、runXxxx 都是没有返回结果的,supplyXxx 都是可以获取返回结果的

    B、可以传入自定义的线程池,否则就用默认的线程池;

    3. 计算完成时回调方法

    在这里插入图片描述

    whenComplete 可以处理正常和异常的计算结果,exceptionally 处理异常情况。

    whenComplete 和 whenCompleteAsync 的区别:

    whenComplete:是执行当前任务的线程执行继续执行 whenComplete 的任务。

    whenCompleteAsync:是执行把 whenCompleteAsync 这个任务继续提交给线程池来进行执行。

    方法不以 Async 结尾,意味着 Action 使用相同的线程执行,而 Async 可能会使用其他线程执行 (如果是使用相同的线程池,也可能会被同一个线程选中执行)。

    public class CompletableFutureDemo {
    	public static void main(String[] args) throws ExecutionException, InterruptedException {
    		CompletableFuture future = CompletableFuture.supplyAsync(new Supplier<Object>() {
    			@Override
    			public Object get() {
    				System.out.println(Thread.currentThread().getName() + "\tcompletableFuture");
    				int i = 10 / 0;
    				return 1024;
    			}
    		}).whenComplete(new BiConsumer<Object, Throwable>() {
    			@Override
    			public void accept(Object o, Throwable throwable) {
    				System.out.println("-------o=" + o.toString());
    				System.out.println("-------throwable=" + throwable);
    			}
    		}).exceptionally(new Function<Throwable, Object>() {
    			@Override
    			public Object apply(Throwable throwable) {
    				System.out.println("throwable=" + throwable);
    				return 6666;
    			}
    		});
    		System.out.println(future.get());
    	}
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25

    4. handle 方法

    在这里插入图片描述

    和 complete 一样,可对结果做最后的处理 (可处理异常),可改变返回值。

    5. 线程串行化方法

    在这里插入图片描述

    thenApply 方法:

    当一个线程依赖另一个线程时,获取上一个任务返回的结果,并返回当前任务的返回值。

    thenAccept 方法:

    消费处理结果。接收任务的处理结果,并消费处理,无返回结果。

    thenRun 方法:

    只要上面的任务执行完成,就开始执行 thenRun,只是处理完任务后,执行thenRun 的后续操作带有 Async 默认是异步执行的。同之前。

    以上都要前置任务成功完成。

    Function<? super T,? extends U>
    T:上一个任务返回结果的类型
    U:当前任务的返回值类型
    
    • 1
    • 2
    • 3

    6. 两任务组合 - 都要完成

    在这里插入图片描述
    两个任务必须都完成,触发该任务。

    thenCombine:

    组合两个 future,获取两个 future 的返回结果,并返回当前任务的返回值

    thenAcceptBoth:

    组合两个 future,获取两个 future 任务的返回结果,然后处理任务,没有返回值。

    runAfterBoth:

    组合两个 future,不需要获取 future 的结果,只需两个future 处理完任务后,处理该任务。

    7. 两任务组合 - 一个完成

    在这里插入图片描述

    当两个任务中,任意一个 future 任务完成的时候,执行任务。

    applyToEither:两个任务有一个执行完成,获取它的返回值,处理任务并有新的返回值。
    acceptEither:两个任务有一个执行完成,获取它的返回值,处理任务,没有新的返回值。runAfterEither:两个任务有一个执行完成,不需要获取 future 的结果,处理任务,也没有返回值。

    8. 多任务组合

    在这里插入图片描述

    allOf:等待所有任务完成
    anyOf:只要有一个任务完成
    
    • 1
    • 2

    9. 示例

    package com.fancy.gulimall.search.thread;
    
    import java.util.concurrent.*;
    
    public class ThreadTest {
        public static ExecutorService executor = Executors.newFixedThreadPool(10);
    
    
        public static void main(String[] args) throws ExecutionException, InterruptedException {
            System.out.println("main....start....");
    //      CompletableFuture future = CompletableFuture.runAsync(() -> {
    //          System.out.println("当前线程:" + Thread.currentThread().getId());
    //          int i = 10 / 2;
    //          System.out.println("运行结果:" + i);
    //      }, executor);
    
            /**
             * 方法完成后的感知
             *          
            */
    //       CompletableFuture future = CompletableFuture.supplyAsync(() -> {
    //            System.out.println("当前线程:" + Thread.currentThread().getId());
    //            int i = 10 / 0;
    //            System.out.println("运行结果:" + i);
    //            return i;
    //       }, executor).whenComplete((res,excption)->{
    //           //虽然能得到异常信息,但是没法修改返回数据。
    //           System.out.println("异步任务成功完成了...结果是:"+res+";异常是:"+excption);
    //       }).exceptionally(throwable -> {
    //           //可以感知异常,同时返回默认值
    //           return 10;
    //       });
    
    
            /**
             * 方法执行完成后的处理
             */
    //       CompletableFuture future = CompletableFuture.supplyAsync(() -> {
    //           System.out.println("当前线程:" + Thread.currentThread().getId());
    //           int i = 10 / 4;
    //           System.out.println("运行结果:" + i);
    //           return i;
    //       }, executor).handle((res, thr) -> {
    //           if (res != null) {
    //               return res * 2;
    //           }
    //           if (thr != null) {
    //               return 0;
    //           }
    //            return 0;
    //       });
            //R apply(T t, U u);
    
            /**
             * 线程串行化
             * 1)、thenRun:不能获取到上一步的执行结果,无返回值
             *  .thenRunAsync(() -> {
             *             System.out.println("任务2启动了...");
             *         }, executor);
             * 2)、thenAcceptAsync;能接受上一步结果,但是无返回值
             * 3)、thenApplyAsync:;能接受上一步结果,有返回值
             */
    //       CompletableFuture future = CompletableFuture.supplyAsync(() -> {
    //           System.out.println("当前线程:" + Thread.currentThread().getId());
    //           int i = 10 / 4;
    //           System.out.println("运行结果:" + i);
    //           return i;
    //       }, executor).thenApplyAsync(res -> {
    //           System.out.println("任务2启动了..." + res);
    //
    //           return "Hello " + res;
    //       }, executor);
            //void accept(T t);
            //R apply(T t);
    
            //future.get()
            /**
             * 两个都完成
             */
    //       CompletableFuture future01 = CompletableFuture.supplyAsync(() -> {
    //           System.out.println("任务1线程:" + Thread.currentThread().getId());
    //           int i = 10 / 4;
    //           System.out.println("任务1结束:" );
    //           return i;
    //       }, executor);
    //
    //       CompletableFuture future02 = CompletableFuture.supplyAsync(() -> {
    //           System.out.println("任务2线程:" + Thread.currentThread().getId());
    //
    //           try {
    //               Thread.sleep(3000);
    //               System.out.println("任务2结束:" );
    //           } catch (InterruptedException e) {
    //               e.printStackTrace();
    //           }
    //           return "Hello";
    //        }, executor);
    
    //     future01.runAfterBothAsync(future02,()->{
    //         System.out.println("任务3开始...");
    //     }, executor);
           // void accept(T t, U u);
    //     future01.thenAcceptBothAsync(future02,(f1,f2)->{
    //         System.out.println("任务3开始...之前的结果:"+f1+"--》"+f2);
    //     }, executor);
           //R apply(T t, U u);
    //     CompletableFuture future = future01.thenCombineAsync(future02, (f1, f2) -> {
    //         return f1 + ":" + f2 + " -> Haha";
    //     }, executor);
    
            /**
             * 两个任务,只要有一个完成,我们就执行任务3
             * runAfterEitherAsync:不感知结果,自己没有返回值
             * acceptEitherAsync:感知结果,自己没有返回值
             * applyToEitherAsync:感知结果,自己有返回值
             */
    //        future01.runAfterEitherAsync(future02,()->{
    //            System.out.println("任务3开始...之前的结果:");
    //        },executor);
            //void accept(T t);
    //        future01.acceptEitherAsync(future02,(res)->{
    //            System.out.println("任务3开始...之前的结果:"+res);
    //        },executor);
    //        CompletableFuture future = future01.applyToEitherAsync(future02, res -> {
    //            System.out.println("任务3开始...之前的结果:" + res);
    //            return res.toString() + "->哈哈";
    //        }, executor);
    
            CompletableFuture<String> futureImg = CompletableFuture.supplyAsync(() -> {
                System.out.println("查询商品的图片信息");
                return "hello.jpg";
            },executor);
    
            CompletableFuture<String> futureAttr = CompletableFuture.supplyAsync(() -> {
                System.out.println("查询商品的属性");
                return "黑色+256G";
            },executor);
    
            CompletableFuture<String> futureDesc = CompletableFuture.supplyAsync(() -> {
                try {
                    Thread.sleep(3000);
                    System.out.println("查询商品介绍");
                } catch (InterruptedException e) {
                    e.printStackTrace();
                }
                return "华为";
            },executor);
    
    //      CompletableFuture allOf = CompletableFuture.allOf(futureImg, futureAttr, futureDesc);
            CompletableFuture<Object> anyOf = CompletableFuture.anyOf(futureImg, futureAttr, futureDesc);
            anyOf.get();//等待所有结果完成
    
    //      System.out.println("main....end...."+futureImg.get()+"=>"+futureAttr.get()+"=>"+futureDesc.get());
            System.out.println("main....end...."+anyOf.get());
    
        }
    }
    
    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25
    • 26
    • 27
    • 28
    • 29
    • 30
    • 31
    • 32
    • 33
    • 34
    • 35
    • 36
    • 37
    • 38
    • 39
    • 40
    • 41
    • 42
    • 43
    • 44
    • 45
    • 46
    • 47
    • 48
    • 49
    • 50
    • 51
    • 52
    • 53
    • 54
    • 55
    • 56
    • 57
    • 58
    • 59
    • 60
    • 61
    • 62
    • 63
    • 64
    • 65
    • 66
    • 67
    • 68
    • 69
    • 70
    • 71
    • 72
    • 73
    • 74
    • 75
    • 76
    • 77
    • 78
    • 79
    • 80
    • 81
    • 82
    • 83
    • 84
    • 85
    • 86
    • 87
    • 88
    • 89
    • 90
    • 91
    • 92
    • 93
    • 94
    • 95
    • 96
    • 97
    • 98
    • 99
    • 100
    • 101
    • 102
    • 103
    • 104
    • 105
    • 106
    • 107
    • 108
    • 109
    • 110
    • 111
    • 112
    • 113
    • 114
    • 115
    • 116
    • 117
    • 118
    • 119
    • 120
    • 121
    • 122
    • 123
    • 124
    • 125
    • 126
    • 127
    • 128
    • 129
    • 130
    • 131
    • 132
    • 133
    • 134
    • 135
    • 136
    • 137
    • 138
    • 139
    • 140
    • 141
    • 142
    • 143
    • 144
    • 145
    • 146
    • 147
    • 148
    • 149
    • 150
    • 151
    • 152
    • 153
    • 154
    • 155
    • 156
    • 157
  • 相关阅读:
    Mysql---三张表(student,课程,分数) 查询课程为数学的学生姓名,编号,成绩
    基于MATLAB的指纹识别算法仿真实现
    AI工程化—— 如何让AI在企业多快好省的落地?
    C++ vector使用方法
    gitee码云的使用
    SEO方案尝试--Nuxtjs项目基础配置
    A Survey on Bias and Fairness in Machine Learning 阅读笔记
    在MuJoCo环境下详细实现PPO算法与Hopper-v2应用教程: 深度学习强化学习实战指南
    C语言—详解库函数中常用的字符串函数
    二叉树的建立和遍历
  • 原文地址:https://blog.csdn.net/m0_51111980/article/details/126876710