Kafka 分布式的，基于发布/订阅的消息系统

Kafka是一种分布式的，基于发布/订阅的消息系统。主要设计目标如下：

• 通过O(1)的磁盘数据结构提供消息的持久化，这种结构对于即使数以TB的消息存储也能够保持长时间的稳定性能。

• 高吞吐量：即使是非常普通的硬件kafka也可以支持每秒数十万的消息。

• Consumer客户端pull，随机读,利用sendfile系统调用进行zero-copy ,批量拉数据

• 消费状态保存在客户端

• 支持Kafka Server间的消息分区，及分布式消费，同时保证每个Partition内的消息顺序传输。

• 数据迁移、扩容对用户透明

• 支持Hadoop并行数据加载。

• 支持online(在线)和offline(离线)的场景。

• 持久化：通过将数据持久化到硬盘以及replication防止数据丢失。

• scale out：无需停机即可扩展机器。

• 定期删除机制，支持设定partitions的segment file保留时间。

项目实例：https://github.com/windwant/kafka-demo

 

kafka.properties

value.serializer=org.apache.kafka.common.serialization.StringSerializer
key.serializer=org.apache.kafka.common.serialization.StringSerializer
request.required.acks=1
bootstrap.servers=localhost:9092

value.deserializer=org.apache.kafka.common.serialization.StringDeserializer
key.deserializer=org.apache.kafka.common.serialization.StringDeserializer
group.id=test-consumer-group

 

Producer:

package org.windwant.kafka;

import org.apache.commons.configuration.ConfigurationException;
import org.apache.commons.configuration.PropertiesConfiguration;
import org.apache.commons.configuration.reloading.FileChangedReloadingStrategy;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.Producer;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.clients.producer.RecordMetadata;

import java.io.IOException;
import java.util.Properties;
import java.util.concurrent.ExecutionException;

/**
 * Producer
 */
public class MyKafkaProducer {

    private Properties props;
    public static void main(String[] args) throws ConfigurationException {
        new MyKafkaProducer().start();
    }

    public MyKafkaProducer() throws ConfigurationException {
        props = new Properties();
        PropertiesConfiguration config = new PropertiesConfiguration("kafka.properties");
        config.setReloadingStrategy(new FileChangedReloadingStrategy());
        //×Ô¶¯±£´æ
        config.setAutoSave(true);
        props.put("value.serializer", config.getString("value.serializer"));
        props.put("key.serializer", config.getString("key.serializer"));
        props.put("request.required.acks", config.getString("request.required.acks"));
        props.put("bootstrap.servers", config.getString("bootstrap.servers"));
    }

    public void start(){
        try {
            Producer<String, String> producer = new KafkaProducer<>(props);
            for(int i = 0; i < 100; i++) {
                RecordMetadata result = producer.send(new ProducerRecord<>("mykafka",
                        "kafka key: " + Integer.toString(i),
                        "kafka value: " + Integer.toString(i))).get();
                System.out.println("producer send: " + result);
                Thread.sleep(1000);
            }
            producer.close();
        } catch (InterruptedException e) {
            e.printStackTrace();
        } catch (ExecutionException e) {
            e.printStackTrace();
        }
    }
}

Consumer:

package org.windwant.kafka;

import org.apache.commons.configuration.ConfigurationException;
import org.apache.commons.configuration.PropertiesConfiguration;
import org.apache.commons.configuration.reloading.FileChangedReloadingStrategy;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.ConsumerRecords;
import org.apache.kafka.clients.consumer.KafkaConsumer;

import java.util.Arrays;
import java.util.Properties;

/**
 * Consumer.
 */
public class MyKafkaConsumer {
    private Properties props;
    public static void main(String[] args) throws ConfigurationException {
        new MyKafkaConsumer().start();
    }

    public MyKafkaConsumer() throws ConfigurationException {
        props = new Properties();
        PropertiesConfiguration config = new PropertiesConfiguration("kafka.properties");
        config.setReloadingStrategy(new FileChangedReloadingStrategy());
        //自动保存
        config.setAutoSave(true);
        props.put("value.deserializer", config.getString("value.deserializer"));
        props.put("key.deserializer", config.getString("key.deserializer"));
        props.put("bootstrap.servers", config.getString("bootstrap.servers"));
        props.put("group.id", config.getString("group.id"));
    }

    public void  start(){
        KafkaConsumer<String, String> consumer = new KafkaConsumer<>(props);
        consumer.subscribe(Arrays.asList("mykafka"));
        while (true) {
            ConsumerRecords<String, String> records = consumer.poll(100);
            for (ConsumerRecord<String, String> record : records) {
                System.out.printf("offset = %d, key = %s, value = %s", record.offset(), record.key(), record.value());
                System.out.println();
            }
        }
    }
}