Disruptor作者，介绍Disruptor能每秒处理600万订单。这是一个可怕的数字。

disruptor之所以那么快，是因为内部采用环形队列和无锁设计。使用cas来进行并发控制。通过获取可用下标来对事件发布和消费

下标通过cas控制(Atomic)

disruptor组成部分

1.Disruptor:用于控制整个消费者-生产者模型的处理器 
　　　2.RingBuffer:用于存放数据 
　　　3.EventHandler:一个用于处理事件的接口（可以当做生产者，也可以当做消费者）。 
　　　4.EventFactory:事件工厂类。 
　　　5.WaitStrategy:用于实现事件处理等待RingBuffer游标策略的接口。 
　　　6.SequeueBarrier:队列屏障，用于处理访问RingBuffer的序列。 
　　　7.用于运行disruptor的线程或者线程池。

Disruptor简单使用

1.创建订单和订单事件

package com.liqang.test;

import java.math.BigDecimal;
/**
 * 简单模拟一个订单
 * @author Administrator
 *
 */
public class Order {
    private int id;
    private BigDecimal price;
    private double num;
    private int pid;
    public int getId() {
        return id;
    }
    public void setId(int id) {
        this.id = id;
    }
    public BigDecimal getPrice() {
        return price;
    }
    public void setPrice(BigDecimal price) {
        this.price = price;
    }
    public double getNum() {
        return num;
    }
    public void setNum(double num) {
        this.num = num;
    }
    public int getPid() {
        return pid;
    }
    public void setPid(int pid) {
        this.pid = pid;
    }
}
package com.liqang.test;

import java.math.BigDecimal;
//订单事件  disruptor容器都是以事件对存在
public class OrderEvent {

    private Order order;

    public Order getOrder() {
        return order;
    }

    public void setOrder(Order order) {
        this.order = order;
    }

}

2.创建disruptor事件工厂

package bhz.base;

import com.lmax.disruptor.EventFactory;
// 需要让disruptor为我们创建事件，我们同时还声明了一个EventFactory来实例化Event对象。
public class LongEventFactory implements EventFactory { 

    @Override
    public Object newInstance() {
        return new LongEvent();
    }
}

disruptor会调用工厂方法为我们创建事件。并放到对应的事件槽里面

3.创建事件消费者处理类

/**
 * 事件消费者
 * @author Administrator
 *
 */
public class OrderEventHandle implements EventHandler<OrderEvent>{

    @Override
    public void onEvent(OrderEvent orderEvent, long arg1, boolean arg2) throws Exception {
        /**
         *做相应的业务处理
         */
        System.out.println(orderEvent.getOrder().getPid());

    }

}

4.创建事件生产者类

/**
 * 事件生产者
 *
 * @author Administrator
 *
 */
public class OrderEvenProducer {
    private RingBuffer<OrderEvent> ringBuffer;// disruptor容器

    public OrderEvenProducer(RingBuffer<OrderEvent> ringBuffer) {
        this.ringBuffer = ringBuffer;
    }

    public void onData(Order order) {
        long index = ringBuffer.next();// 首先获取下一个事件槽位置
        try {
            OrderEvent orderEvent = ringBuffer.get(index);// 通过序列获得disruptorFacotry创建好的事件槽
            orderEvent.setOrder(order);// 填充好业务数据
        } finally {
            ringBuffer.publish(index);// 发布事件。使用finally保证publish调用
        }

    }
}

任务是 根据容器 往容器里面注册数据

/**
 * 事件生产者
 *
 * @author Administrator
 *
 */
public class OrderEvenProducer {
    private RingBuffer<OrderEvent> ringBuffer;// disruptor容器

    public OrderEvenProducer(RingBuffer<OrderEvent> ringBuffer) {
        this.ringBuffer = ringBuffer;
    }

    public void onData(Order order) {
        long index = ringBuffer.next();// 首先获取下一个事件槽位置
        try {
            OrderEvent orderEvent = ringBuffer.get(index);// 通过序列获得disruptorFacotry创建好的事件槽
            orderEvent.setOrder(order);// 填充好业务数据
        } finally {
            ringBuffer.publish(index);// 发布事件。使用finally保证publish调用
        }

    }
}

5.测试

package com.liqang.test;

import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

import com.lmax.disruptor.YieldingWaitStrategy;
import com.lmax.disruptor.dsl.Disruptor;
import com.lmax.disruptor.dsl.ProducerType;

import bhz.base.LongEventHandler;

public class OrderEventMain {
    public static void main(String[] args) {
        /**
         * 创建线程池 不限制大小 60秒不被使用就会被回收
         */
        ExecutorService executorService = Executors.newCachedThreadPool();

        OrderEventFactory factory = new OrderEventFactory();// 创建事件工厂
        // 创建bufferSize ,也就是RingBuffer大小，必须是2的N次方 diruptor减少计算事件槽的时间
        int ringBufferSize = 1024 * 1024; //
        /**
         * //BlockingWaitStrategy 是最低效的策略，但其对CPU的消耗最小并且在各种不同部署环境中能提供更加一致的性能表现
         * WaitStrategy BLOCKING_WAIT = new BlockingWaitStrategy();
         * //SleepingWaitStrategy
         * 的性能表现跟BlockingWaitStrategy差不多，对CPU的消耗也类似，但其对生产者线程的影响最小，适合用于异步日志类似的场景
         * WaitStrategy SLEEPING_WAIT = new SleepingWaitStrategy();
         * //YieldingWaitStrategy
         * 的性能是最好的，适合用于低延迟的系统。在要求极高性能且事件处理线数小于CPU逻辑核心数的场景中，推荐使用此策略；例如，CPU开启超线程的特性
         * WaitStrategy YIELDING_WAIT = new YieldingWaitStrategy();
         */
        // 初始化disruptorProducerType.SINGLE 表示是单生产者
        Disruptor<OrderEvent> disruptor = new Disruptor<>(factory, ringBufferSize, executorService, ProducerType.SINGLE,
                new YieldingWaitStrategy());
        // 注册消费者事件处理器
        disruptor.handleEventsWith(new OrderEventHandle());
        // 启动
        disruptor.start();
        //创建生产者
        OrderEvenProducer orderEvenProducer=new OrderEvenProducer(disruptor.getRingBuffer());

        //模拟生产10个订单
        for (int i = 0; i <10; i++) {
            Order order=new Order();
            order.setId(i);
            order.setNum(i);
            order.setPid(i);
            order.setPid(i);
            orderEvenProducer.onData(order);
        }
        disruptor.shutdown();//关闭 disruptor，方法会堵塞，直至所有的事件都得到处理；
        executorService.shutdown();//关闭 disruptor 使用的线程池；如果需要的话，必须手动关闭， disruptor 在 shutdown 时不会自动关闭；
    }
}

disruptor3.0提供lambda表达式的方式(需要jdk8)发布事件 改造事件发布者类

package com.liqang.test;

import java.nio.ByteBuffer;

import com.lmax.disruptor.EventTranslatorOneArg;
import com.lmax.disruptor.RingBuffer;

import bhz.base.LongEvent;

/**
 * 事件生产者
 *
 * @author Administrator
 *
 */
public class OrderEvenProducer {
    private static final EventTranslatorOneArg<OrderEvent, Order> TRANSLATOR =
            new EventTranslatorOneArg<OrderEvent, Order>() {
                @Override
                public void translateTo(OrderEvent event, long sequeue, Order order) {
                    event.setOrder(order);
                }
            };

    private final RingBuffer<OrderEvent> ringBuffer;

    public OrderEvenProducer(RingBuffer<OrderEvent> ringBuffer) {
        this.ringBuffer = ringBuffer;
    }

    public void onData(Order order){
        ringBuffer.publishEvent(TRANSLATOR, order);

        //ringBuffer.publishEvent((event,sequeue,or)->{event.setOrder(order);},order);
    }
}

直接使用RingBuffer

package com.liqang.test;

import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

import com.lmax.disruptor.BatchEventProcessor;
import com.lmax.disruptor.EventFactory;
import com.lmax.disruptor.RingBuffer;
import com.lmax.disruptor.SequenceBarrier;
import com.lmax.disruptor.YieldingWaitStrategy;

import bhz.generate1.Trade;
import bhz.generate1.TradeHandler;

public class RingBufferTest {
    public static void main(String[] args) throws InterruptedException, ExecutionException {

     int bufferSize=1024*1024;
     //创建线程池
     ExecutorService executorService=Executors.newCachedThreadPool();
     //初始化ringbuffer
      RingBuffer<OrderEvent> ringBuffer=RingBuffer.createSingleProducer(new EventFactory<OrderEvent>() {

        @Override
        public OrderEvent newInstance() {
            // TODO Auto-generated method stub
            return new OrderEvent();
        }
    }, bufferSize, new YieldingWaitStrategy());
      //创建SequenceBarrier
      SequenceBarrier sequenceBarrier = ringBuffer.newBarrier();  

      //创建消息处理器
      BatchEventProcessor<OrderEvent> transProcessor = new BatchEventProcessor<OrderEvent>(
              ringBuffer, sequenceBarrier, new OrderEventHandle());  

      //这一步的目的就是把消费者的位置信息引用注入到生产者    如果只有一个消费者的情况可以省略
      ringBuffer.addGatingSequences(transProcessor.getSequence());  

      //把消息处理器提交到线程池
      executorService.submit(transProcessor);  

      //如果存在多个消费者 那重复执行上面3行代码 把TradeHandler换成其它消费者类  

      Future<?> future= executorService.submit(new Callable<Void>() {
          @Override
          public Void call() throws Exception {
              long seq;
              for(int i=0;i<10;i++){
                  seq = ringBuffer.next();//占个坑 --ringBuffer一个可用区块
                  OrderEvent orderEvent=ringBuffer.get(seq);
                  Order order=new Order();
                  order.setId(i);
                  order.setNum(i);
                  order.setPid(i);
                  order.setPid(i);
                  orderEvent.setOrder(order);
                  ringBuffer.publish(seq);//发布这个区块的数据使handler(consumer)可见
              }
              return null;
          }
      }); 

      future.get();//等待生产者结束
      Thread.sleep(1000);//等上1秒，等消费都处理完成
      transProcessor.halt();//通知事件(或者说消息)处理器 可以结束了（并不是马上结束!!!）
      executorService.shutdown();//终止线程
}
}

Disruptor做复杂业务操作

disruptor还可以做很多复杂的业务操作

如 一个事件c1 c2 处理器并行执行 执行完之后交给c3

public class Main {
    public static void main(String[] args) throws InterruptedException {  

        long beginTime=System.currentTimeMillis();
        int bufferSize=1024;
        ExecutorService executor=Executors.newFixedThreadPool(8);  

        Disruptor<Trade> disruptor = new Disruptor<Trade>(new EventFactory<Trade>() {
            @Override
            public Trade newInstance() {
                return new Trade();
            }
        }, bufferSize, executor, ProducerType.SINGLE, new BusySpinWaitStrategy());  

        //菱形操作

        //使用disruptor创建消费者组C1,C2
        EventHandlerGroup<Trade> handlerGroup =
                disruptor.handleEventsWith(new Handler1(), new Handler2());
        //声明在C1,C2完事之后执行JMS消息发送操作 也就是流程走到C3 (测试遇到一个问题就是 要队列被消费完了才会走到3)
        handlerGroup.then(new Handler3());

        disruptor.start();//启动
        CountDownLatch latch=new CountDownLatch(1);
        //生产者准备
        executor.submit(new TradePublisher(latch, disruptor));

        latch.await();//等待生产者完事. 

        disruptor.shutdown();
        executor.shutdown();
        System.out.println("总耗时:"+(System.currentTimeMillis()-beginTime));
    }
}  
package bhz.generate2;

import java.util.Random;
import java.util.concurrent.CountDownLatch;

import bhz.generate1.Trade;

import com.lmax.disruptor.EventTranslator;
import com.lmax.disruptor.dsl.Disruptor;

public class TradePublisher implements Runnable {  

    Disruptor<Trade> disruptor;
    private CountDownLatch latch;  

    private static int LOOP=1000;//模拟百万次交易的发生  

    public TradePublisher(CountDownLatch latch,Disruptor<Trade> disruptor) {
        this.disruptor=disruptor;
        this.latch=latch;
    }  

    @Override
    public void run() {  for(int i=0;i<LOOP;i++){
            disruptor.getRingBuffer().publishEvent((event,sequeue,or)->{event.setId("ff");},new Trade());
        }
        latch.countDown();
    }  

}

顺序执行

c1执行后交给c2处理 c2 处理后交给c3处理

 package bhz.generate2;

import java.util.concurrent.CountDownLatch;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

import bhz.generate1.Trade;
import bhz.generate1.TradeHandler;

import com.lmax.disruptor.BusySpinWaitStrategy;
import com.lmax.disruptor.EventFactory;
import com.lmax.disruptor.dsl.Disruptor;
import com.lmax.disruptor.dsl.EventHandlerGroup;
import com.lmax.disruptor.dsl.ProducerType;

public class Main {
    public static void main(String[] args) throws InterruptedException {  

        long beginTime=System.currentTimeMillis();
        int bufferSize=1024;
        ExecutorService executor=Executors.newFixedThreadPool(8);  

        Disruptor<Trade> disruptor = new Disruptor<Trade>(new EventFactory<Trade>() {
            @Override
            public Trade newInstance() {
                return new Trade();
            }
        }, bufferSize, executor, ProducerType.SINGLE, new BusySpinWaitStrategy());  

        //顺序操作

        disruptor.handleEventsWith(new Handler1()).
            handleEventsWith(new Handler2()).
            handleEventsWith(new Handler3());

        disruptor.start();//启动
        CountDownLatch latch=new CountDownLatch(1);
        //生产者准备
        executor.submit(new TradePublisher(latch, disruptor));

        latch.await();//等待生产者完事. 

        disruptor.shutdown();
        executor.shutdown();
        System.out.println("总耗时:"+(System.currentTimeMillis()-beginTime));
    }
}

六边形操作

 package bhz.generate2;

import java.util.concurrent.CountDownLatch;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;

import bhz.generate1.Trade;
import bhz.generate1.TradeHandler;

import com.lmax.disruptor.BusySpinWaitStrategy;
import com.lmax.disruptor.EventFactory;
import com.lmax.disruptor.dsl.Disruptor;
import com.lmax.disruptor.dsl.EventHandlerGroup;
import com.lmax.disruptor.dsl.ProducerType;

public class Main {
    public static void main(String[] args) throws InterruptedException {  

        long beginTime=System.currentTimeMillis();
        int bufferSize=1024;
        ExecutorService executor=Executors.newFixedThreadPool(8);  

        Disruptor<Trade> disruptor = new Disruptor<Trade>(new EventFactory<Trade>() {
            @Override
            public Trade newInstance() {
                return new Trade();
            }
        }, bufferSize, executor, ProducerType.SINGLE, new BusySpinWaitStrategy());  

        //六边形操作. 

        Handler1 h1 = new Handler1();
        Handler2 h2 = new Handler2();
        Handler3 h3 = new Handler3();
        Handler4 h4 = new Handler4();
        Handler5 h5 = new Handler5();
        disruptor.handleEventsWith(h1, h2);
        disruptor.after(h1).handleEventsWith(h4);
        disruptor.after(h2).handleEventsWith(h5);
        disruptor.after(h4, h5).handleEventsWith(h3);

        disruptor.start();//启动
        CountDownLatch latch=new CountDownLatch(1);
        //生产者准备
        executor.submit(new TradePublisher(latch, disruptor));

        latch.await();//等待生产者完事. 

        disruptor.shutdown();
        executor.shutdown();
        System.out.println("总耗时:"+(System.currentTimeMillis()-beginTime));
    }
}

h1 h2并行执行  h1执行完毕之后h4执行 h2执行完毕之后h5执行最终交给h3汇总执行

Disruptor使用的更多相关文章

  1. 架构师养成记--15.Disruptor并发框架

    一.概述 disruptor对于处理并发任务很擅长,曾有人测过,一个线程里1s内可以处理六百万个订单,性能相当感人. 这个框架的结构大概是:数据生产端 --> 缓存 --> 消费端 缓存中 ...

  2. 并发框架Disruptor浅析

    1.引言 Disruptor是一个开源的Java框架,它被设计用于在生产者—消费者(producer-consumer problem,简称PCP)问题上获得尽量高的吞吐量(TPS)和尽量低的延迟.D ...

  3. LMAX Disruptor—多生产者多消费者中，消息复制分发的高性能实现

    解决的问题 当我们有多个消息的生产者线程,一个消费者线程时,他们之间如何进行高并发.线程安全的协调? 很简单,用一个队列. 当我们有多个消息的生产者线程,多个消费者线程,并且每一条消息需要被所有的消费 ...

  4. Disruptor 极速体验

    已经不记得最早接触到 Disruptor 是什么时候了,只记得发现它的时候它是以具有闪电般的速度被介绍的.于是在脑子里, Disruptor 和"闪电"一词关联了起来,然而却一直没 ...

  5. disruptor - Concurrent Programming Framework 并发编程框架

    disruptor发布了Java的2.0版本(.Net版本见这里),disruptor是一个高性能的异步处理框架,或者可以认为是最快的消息框架(轻量的JMS),也可以认为是一个观察者模式实现,或者事件 ...

  6. 剖析Disruptor:为什么会这么快？（二）神奇的缓存行填充

    原文链接:http://mechanitis.blogspot.com/2011/07/dissecting-disruptor-why-its-so-fast_22.html 需FQ 计算机入门   ...

  7. The LMAX disruptor Architecture--转载

    原文地址: LMAX is a new retail financial trading platform. As a result it has to process many trades wit ...

  8. Disruptor 源码阅读笔记--转

    原文地址:http://coderbee.net/index.php/open-source/20130812/400 一.Disruptor 是什么? Disruptor 是一个高性能异步处理框架, ...

  9. 构建高性能服务（三）Java高性能缓冲设计 vs Disruptor vs LinkedBlockingQueue--转载

    原文地址:http://maoyidao.iteye.com/blog/1663193 一个仅仅部署在4台服务器上的服务,每秒向Database写入数据超过100万行数据,每分钟产生超过1G的数据.而 ...

  10. LMAX Disruptor – High Performance, Low Latency and Simple Too 转载

    原文地址:http://www.symphonious.net/2011/07/11/lmax-disruptor-high-performance-low-latency-and-simple-to ...

随机推荐

  1. silverlight学习笔记——新手对silverlight的认识（1）

    这几天在搞silverlight.虽然silverlight没有前途,但始终是微软的一门技术,界面基本上与WPF共通,用一下也无妨. 学习过程并没有我原先想得那么容易,有些地方捣鼓了很久.究其原因,是 ...

  2. 【数据结构】链式向前星知识点&代码

    代码: struct NODE{ int to; int nxt; int c; }node[MM];//链式向前星 ; void add(int a,int b,int c){ node[lcnt] ...

  3. C#如何读写和创建INI文件（经典）转

    C#如何读写和创建INI文件 分类: c#程序设计2011-11-27 20:42 4935人阅读 评论(2) 收藏 举报 inic#stringbuffernullfile 在做项目过程中,有时需要 ...

  4. 第2课 Git配置文件的妙用

    2-1 "git config"  指令的用法 文件夹中".git"子文件夹内的config文件的优先权>登录账号的home directory中的.gi ...

  5. Genesis 多边形闭轮廓填充算法

    通过逐行扫描,计算得出直线与多边形相交点进行求解 原理图形如下所示: 相关函数: /// <summary> /// 求点P到线段L距离 /// </summary> /// ...

  6. php 提交编辑数据没有变,返回0,判断

    php 提交编辑数据没有变,返回0,判断以TP为例子 $edit = D('Brand')->save($data);if($edit == true){ echo "修改成功&quo ...

  7. Aspnet_Session

    cmd: aspnet_regsql.exe -ssadd -sstype c -d ZZCasSession -S 192.168.0.3 -U sa -P szhweb2010 <!--会话 ...

  8. 关于TJOI2014的一道题——Alice and Bob

    B Alice and Bob •输入输出文件: alice.in/alice.out •源文件名: alice.cpp/alice.c/alice.pas • 时间限制: 1s 内存限制: 128M ...

  9. ACM_魔仙岛探险（深搜）

    魔仙岛探险 Time Limit: 2000/1000ms (Java/Others) Problem Description: 小敏通过秘密方法得到一张不完整的魔仙岛航拍地图.魔仙岛由一个主岛和一些 ...

  10. myeclipse2014 破解步骤

    1.打开破解文件夹Myeclipse 2014 patch,运行run.bat文件 2.在破解界面中,usercode随便输入, systemid 在右边的SystemId按钮处,点击. 将自动生成一 ...