Disruptor完成多个消费者不重复消费消息
Disruptor完成多个消费者并行、顺序重复消费Event。重复消费类似于kafka中,同一个topic被不同的group的消费者消费。这样的场景比较常见。当然更常见的场景是不重复消费,也就是一个消息只能被消费一次。
Disruptor同样可以完成不重复消费的功能。
上一篇消费者消费Handler是这样的,需要实现EventHandler。
/**
-
@author wuweifeng wrote on 2018/3/29.
*/
public class LastEventHandler implements EventHandler {@Override
public void onEvent(LongEvent secondEvent, long sequence, boolean endOfBatch) throws Exception {
System.out.println("-------Last:" + secondEvent);
}
}
这次的消费者需要实现不同的接口,代码如下,注意,代码是接续上一篇的,需要先下载上一篇的代码。
import a.LongEvent;
import com.lmax.disruptor.WorkHandler;
/**
- @author wuweifeng wrote on 2018/4/8.
*/
public class Consumer implements WorkHandler {
@Override
public void onEvent(LongEvent longEvent) throws Exception {
System.out.println(Thread.currentThread().getName() + “消费者消费了消息:” + longEvent.toString());
}
}
import a.LongEvent;
import a.LongEventFactory;
import a.LongEventProducer;
import com.lmax.disruptor.BlockingWaitStrategy;
import com.lmax.disruptor.RingBuffer;
import com.lmax.disruptor.dsl.Disruptor;
import com.lmax.disruptor.dsl.ProducerType;
import java.nio.ByteBuffer;
import java.util.concurrent.Executors;
import java.util.concurrent.ThreadFactory;
/**
-
@author wuweifeng wrote on 2018/4/8.
*/
@SuppressWarnings(“ALL”)
public class Test {
public static void main(String[] args) {
ThreadFactory producerFactory = Executors.defaultThreadFactory();
// 创建缓冲池
LongEventFactory eventFactory = new LongEventFactory();// 创建bufferSize ,也就是RingBuffer大小,必须是2的N次方 int ringBufferSize = 1024 * 1024; Disruptor<LongEvent> disruptor = new Disruptor<>(eventFactory, ringBufferSize, producerFactory, ProducerType.SINGLE, new BlockingWaitStrategy()); RingBuffer<LongEvent> ringBuffer = disruptor.getRingBuffer(); // 创建10个消费者来处理同一个生产者发的消息(这10个消费者不重复消费消息) Consumer[] consumers = new Consumer[10]; for (int i = 0; i < consumers.length; i++) { consumers[i] = new Consumer(); } disruptor.handleEventsWithWorkerPool(consumers); disruptor.start(); LongEventProducer longEventProducer = new LongEventProducer(ringBuffer); ByteBuffer bb = ByteBuffer.allocate(8); for (long i = 0; i < 100L; i++) { bb.putLong(0, i); longEventProducer.onData(bb); } disruptor.shutdown();
}
}
这个是测试类,可以发现和上一篇的主要区别就是disruptor.handleEventsWithWorkerPool。其他的都是一样的。
使用handleEventsWithWorkerPool就可以完成不重复消费,使用handleEventsWith就是重复消费。这里定义了10个消费者,那么就会启动10个线程来不重复消费生产者发出的100条消息。