Map(new TopicPartition(topic, partition) -> 2L)
val stream = KafkaUtils.createDirectStream[String, String](
ssc,
PreferConsistent,
Subscribe[String, String](topics, kafkaParams,offsets))
override def partition(topic: String, key: Any, keyBytes: Array[Byte], value: Any, valueBytes: Array[Byte], cluster: Cluster): Int = {
var partition = 0
val keyInt = Integer.parseInt(key.asInstanceOf[String])
val tripData = value.asInstanceOf[String]
//Gets the UserType from the message produced
val userType = tripData.split(",")(12)
//Assigns the partitions to the messages based on the user types
if ("Subscriber".equalsIgnoreCase(userType)) {
partition = 0;
} else if ("Customer".equalsIgnoreCase(userType)) {
partition = 1;
}
println("Partition for message " + value + " is " + partition)
partition
}
import kafka.producer._
import kafka.utils._
import java.util._
import java.text._
import java.util.concurrent.atomic._
class KafkaPartitioner(props: VerifiableProperties = null) extends Partitioner {
val counter = new AtomicInteger(0)
val batch = new AtomicInteger(0)
val partition = new AtomicInteger(0)
def partition(key: Any, numPartitions: Int): Int = {
//round robin partitioner to smooth producers' traffic on all partitions
//change partition every X messages where X corresponds to kafka producer batch message size
if(batch.incrementAndGet % BATCH_NUM_MESSAGES == 1) {
partition.set(math.abs(counter.incrementAndGet) % numPartitions)
batch.set(0)
}
partition.get
}
}
import java.util.*;
import org.apache.kafka.clients.producer.*;
public class SensorProducer {
public static void main(String[] args) throws Exception{
String topicName = "SensorTopic";
Properties props = new Properties();
props.put("bootstrap.servers", "localhost:9092,localhost:9093");
props.put("key.serializer","org.apache.kafka.common.serialization.StringSerializer");
props.put("value.serializer", "org.apache.kafka.common.serialization.StringSerializer");
props.put("partitioner.class", "SensorPartitioner");
props.put("speed.sensor.name", "TSS");
Producer<String, String> producer = new KafkaProducer<>(props);
for (inti=0 ;i<10 ; i++)
producer.send(new ProducerRecord<>(topicName,"SSP"+i,"500"+i));
for (inti=0 ;i<10 ; i++)
producer.send(new ProducerRecord<>(topicName,"TSS","500"+i));
producer.close();
System.out.println("SimpleProducer Completed.");
}
}
import java.util.*;
import org.apache.kafka.clients.producer.*;
import org.apache.kafka.common.*;
import org.apache.kafka.common.utils.*;
import org.apache.kafka.common.record.*;
public class SensorPartitioner implements Partitioner {
private String speedSensorName;
public void configure(Map<String, ?> configs) {
speedSensorName=configs.get("speed.sensor.name").toString();
}
public intpartition(String topic, Object key, byte[] keyBytes, Object value, byte[] valueBytes, Cluster cluster) {
List<PartitionInfo> partitions =cluster.partitionsForTopic(topic);
intnumPartitions = partitions.size();
intsp = (int)Math.abs(numPartitions*0.3);
int p=0;
if ( (keyBytes == null) || (!(key instanceof String)) )
throw new InvalidRecordException("All messages must have sensor name as key");
if ( ((String)key).equals(speedSensorName) )
p = Utils.toPositive(Utils.murmur2(valueBytes)) % sp;
else
p = Utils.toPositive(Utils.murmur2(keyBytes)) % (numPartitions-sp) + sp ;
System.out.println("Key = " + (String)key + " Partition = " + p );
return p;
}
public void close() {}
}
Great Article
ReplyDeleteIEEE Projects for CSE in Big Data
Java Training in Chennai
Final Year Project Centers in Chennai
Java Training in Chennai
ataşehir samsung klima servisi
ReplyDeleteçekmeköy daikin klima servisi
ataşehir daikin klima servisi
maltepe toshiba klima servisi
kadıköy toshiba klima servisi
maltepe beko klima servisi
kadıköy beko klima servisi
kartal lg klima servisi
ümraniye lg klima servisi