美文网首页Dubbo源码解析线程那些事Springboot
线程池那些事之Dubbo线程池设计

线程池那些事之Dubbo线程池设计

作者: 土豆肉丝盖浇饭 | 来源:发表于2018-11-07 00:27 被阅读2次

    前言

    Dubbo的IO模型中提供了4种线程池,下面我会一一介绍。推荐先阅读我前几篇的线程池解析,更有助于理解Dubbo中的线程池设计。

    在Dubbo中什么时候会用到线程池

    图片上看起来只有服务端用到了ThreadPool,实际上客户端调用的时候也会用到ThreadPool

    Dubbo的线程池扩展用作业务线程池,在Dubbo的Dispatcher模块会用到这些线程池,Dispatcher这个模块用于决定Netty ChannelHandler 那些事件的处理需要在业务线程池执行,对于那些耗时的阶段推荐在业务线程池运行,而不是在IO线程中,Dubbo中默认使用的线程池为LimitedThreadPool。

    下面先看下Dubbo提供的4种线程池的源码

    源码解析

    CachedThreadPool

    public class CachedThreadPool implements ThreadPool {
    
        @Override
        public Executor getExecutor(URL url) {
            String name = url.getParameter(Constants.THREAD_NAME_KEY, Constants.DEFAULT_THREAD_NAME);
            int cores = url.getParameter(Constants.CORE_THREADS_KEY, Constants.DEFAULT_CORE_THREADS);
            int threads = url.getParameter(Constants.THREADS_KEY, Integer.MAX_VALUE);
            int queues = url.getParameter(Constants.QUEUES_KEY, Constants.DEFAULT_QUEUES);
            int alive = url.getParameter(Constants.ALIVE_KEY, Constants.DEFAULT_ALIVE);
            return new ThreadPoolExecutor(cores, threads, alive, TimeUnit.MILLISECONDS,
                    queues == 0 ? new SynchronousQueue<Runnable>() :
                            (queues < 0 ? new LinkedBlockingQueue<Runnable>()
                                    : new LinkedBlockingQueue<Runnable>(queues)),
                    new NamedInternalThreadFactory(name, true), new AbortPolicyWithReport(name, url));
        }
    }
    

    缓冲线程池,默认配置如下

    配置 配置值
    corePoolSize 0
    maximumPoolSize Integer.MAX_VALUE
    keepAliveTime 60s
    workQueue 根据queue决定是SynchronousQueue还是LinkedBlockingQueue,默认queue=0,所以是SynchronousQueue
    threadFactory NamedInternalThreadFactory
    rejectHandler AbortPolicyWithReport

    就默认配置来看,和Executors创建的差不多,存在内存溢出风险。NamedInternalThreadFactory主要用于修改线程名,方便我们排查问题。AbortPolicyWithReport对拒绝的任务打印日志,也是方便排查问题。

    LimitedThreadPool

    public class LimitedThreadPool implements ThreadPool {
    
        @Override
        public Executor getExecutor(URL url) {
            String name = url.getParameter(Constants.THREAD_NAME_KEY, Constants.DEFAULT_THREAD_NAME);
            int cores = url.getParameter(Constants.CORE_THREADS_KEY, Constants.DEFAULT_CORE_THREADS);
            int threads = url.getParameter(Constants.THREADS_KEY, Constants.DEFAULT_THREADS);
            int queues = url.getParameter(Constants.QUEUES_KEY, Constants.DEFAULT_QUEUES);
            return new ThreadPoolExecutor(cores, threads, Long.MAX_VALUE, TimeUnit.MILLISECONDS,
                    queues == 0 ? new SynchronousQueue<Runnable>() :
                            (queues < 0 ? new LinkedBlockingQueue<Runnable>()
                                    : new LinkedBlockingQueue<Runnable>(queues)),
                    new NamedInternalThreadFactory(name, true), new AbortPolicyWithReport(name, url));
        }
    
    }
    
    配置 配置值
    corePoolSize 0
    maximumPoolSize 200
    keepAliveTime Long.MAX_VALUE,相当于无限长
    workQueue 根据queue决定是SynchronousQueue还是LinkedBlockingQueue,默认queue=0,所以是SynchronousQueue
    threadFactory NamedInternalThreadFactory
    rejectHandler AbortPolicyWithReport

    从keepAliveTime的配置可以看出来,LimitedThreadPool线程池的特性是线程数只会增加不会减少。

    FixedThreadPool

    public class FixedThreadPool implements ThreadPool {
    
        @Override
        public Executor getExecutor(URL url) {
            String name = url.getParameter(Constants.THREAD_NAME_KEY, Constants.DEFAULT_THREAD_NAME);
            int threads = url.getParameter(Constants.THREADS_KEY, Constants.DEFAULT_THREADS);
            int queues = url.getParameter(Constants.QUEUES_KEY, Constants.DEFAULT_QUEUES);
            return new ThreadPoolExecutor(threads, threads, 0, TimeUnit.MILLISECONDS,
                    queues == 0 ? new SynchronousQueue<Runnable>() :
                            (queues < 0 ? new LinkedBlockingQueue<Runnable>()
                                    : new LinkedBlockingQueue<Runnable>(queues)),
                    new NamedInternalThreadFactory(name, true), new AbortPolicyWithReport(name, url));
        }
    
    }
    
    配置 配置值
    corePoolSize 200
    maximumPoolSize 200
    keepAliveTime 0
    workQueue 根据queue决定是SynchronousQueue还是LinkedBlockingQueue,默认queue=0,所以是SynchronousQueue
    threadFactory NamedInternalThreadFactory
    rejectHandler AbortPolicyWithReport

    Dubbo的默认线程池,固定200个线程,就配置来看和LimitedThreadPool没什么区别,因为这两种线程池都是只增加。

    EagerThreadPool

    public class EagerThreadPool implements ThreadPool {
    
        @Override
        public Executor getExecutor(URL url) {
            String name = url.getParameter(Constants.THREAD_NAME_KEY, Constants.DEFAULT_THREAD_NAME);
            int cores = url.getParameter(Constants.CORE_THREADS_KEY, Constants.DEFAULT_CORE_THREADS);
            int threads = url.getParameter(Constants.THREADS_KEY, Integer.MAX_VALUE);
            int queues = url.getParameter(Constants.QUEUES_KEY, Constants.DEFAULT_QUEUES);
            int alive = url.getParameter(Constants.ALIVE_KEY, Constants.DEFAULT_ALIVE);
    
            // init queue and executor
            TaskQueue<Runnable> taskQueue = new TaskQueue<Runnable>(queues <= 0 ? 1 : queues);
            EagerThreadPoolExecutor executor = new EagerThreadPoolExecutor(cores,
                    threads,
                    alive,
                    TimeUnit.MILLISECONDS,
                    taskQueue,
                    new NamedInternalThreadFactory(name, true),
                    new AbortPolicyWithReport(name, url));
            taskQueue.setExecutor(executor);
            return executor;
        }
    }
    
    配置 配置值
    corePoolSize 0
    maximumPoolSize Integer.MAX_VALUE
    keepAliveTime 60s
    workQueue 自定义实现TaskQueue,默认长度为1,使用时要自己配置下
    threadFactory NamedInternalThreadFactory
    rejectHandler AbortPolicyWithReport

    我们知道,当线程数量达到corePoolSize之后,只有当workqueue满了之后,才会增加工作线程。
    这个线程池就是对这个特性做了优化,首先继承ThreadPoolExecutor实现EagerThreadPoolExecutor,对当前线程池提交的任务数submittedTaskCount进行记录。
    其次是通过自定义TaskQueue作为workQueue,它会在提交任务时判断是否currentPoolSize<submittedTaskCount<maxPoolSize,然后通过它的offer方法返回false导致增加工作线程。

    为什么返回false会增加工作线程,我们回顾下ThreadPoolExecutor的execute方法

    public void execute(Runnable command) {
            if (command == null)
                throw new NullPointerException();
            int c = ctl.get();
            if (workerCountOf(c) < corePoolSize) {
                if (addWorker(command, true))
                    return;
                c = ctl.get();
            }
            //在这一步如果offer方法返回false,那么会进入到下一个else分支判断
            //如果这个时候当前工作线程数没有达到上限,那么就会增加一个工作线程
            //对于普通的workQueue在没有满的情况下是不会返回false的
            if (isRunning(c) && workQueue.offer(command)) {
                int recheck = ctl.get();
                if (! isRunning(recheck) && remove(command))
                    reject(command);
                else if (workerCountOf(recheck) == 0)
                    addWorker(null, false);
            }
            else if (!addWorker(command, false))
                reject(command);
        }
    

    然后看下TaskQueue的offer方法逻辑

     public boolean offer(Runnable runnable) {
            //TaskQueue持有executor引用,用于获取当前提交任务数
            if (executor == null) {
                throw new RejectedExecutionException("The task queue does not have executor!");
            }
    
            int currentPoolThreadSize = executor.getPoolSize();
            //如果提交任务数小于当前工作线程数,说明当前工作线程足够处理任务,将提交的任务插入到工作队列
            if (executor.getSubmittedTaskCount() < currentPoolThreadSize) {
                return super.offer(runnable);
            }
    
            //如果提交任务数大于当前工作线程数并且小于最大线程数,说明提交的任务量线程已经处理不过来,那么需要增加线程数,返回false
            if (currentPoolThreadSize < executor.getMaximumPoolSize()) {
                return false;
            }
    
            //工作线程数到达最大线程数,插入到workqueue
            return super.offer(runnable);
        }
    

    再看下修改的EagerThreadPoolExecutor如何统计提交的任务

        public int getSubmittedTaskCount() {
            return submittedTaskCount.get();
        }
    
        //任务执行完毕后,submittedTaskCount--
        @Override
        protected void afterExecute(Runnable r, Throwable t) {
            submittedTaskCount.decrementAndGet();
        }
    
        @Override
        public void execute(Runnable command) {
            if (command == null) {
                throw new NullPointerException();
            }
            //执行任务前,submittedTaskCount++
            submittedTaskCount.incrementAndGet();
            try {
                super.execute(command);
            } catch (RejectedExecutionException rx) {
               //失败尝试重新加入到workqueue
                final TaskQueue queue = (TaskQueue) super.getQueue();
                try {
                    if (!queue.retryOffer(command, 0, TimeUnit.MILLISECONDS)) {
                        //如果重新加入失败,那么抛出异常,并且统计数-1
                        submittedTaskCount.decrementAndGet();
                        throw new RejectedExecutionException("Queue capacity is full.");
                    }
                } catch (InterruptedException x) {
                    submittedTaskCount.decrementAndGet();
                    throw new RejectedExecutionException(x);
                }
            } catch (Throwable t) {
                // decrease any way
                submittedTaskCount.decrementAndGet();
            }
        }
    

    最后

    一般来讲使用Dubbo的默认配置,我们公司的业务量还没到需要对线程池进行特殊配置的地步。本文主要目的是,通过一个成熟框架对线程池的配置点,指导我们在实际使用线程池中需要注意的点。

    下面是我公众号,大家可以关注下。


    image

    相关文章

      网友评论

        本文标题:线程池那些事之Dubbo线程池设计

        本文链接:https://www.haomeiwen.com/subject/bgilaftx.html