Skip to content

Latest commit

 

History

History
142 lines (101 loc) · 4.88 KB

README.md

File metadata and controls

142 lines (101 loc) · 4.88 KB
Raw

Universidad EAFIT

Curso ST0237 Big Data 20181-1

Profesor: Edwin Montoya M. – [email protected]

MapReduce

(1) WordCount en Java

Tomado de: https://hadoop.apache.org/docs/r2.7.3/hadoop-mapreduce-client/hadoop-mapreduce-client-core/MapReduceTutorial.html

  • Contador de palabras en archivos texto en Java

  • Se tiene el programa ejemplo: WordCount.java, el cual despues de compilarse en la version hadoop 2.7.3, genera un jar wc.jar, el cual será el que finalmente se ejecute.

  • Descargarlo, compilarlo y generar el jar (wc.jar)

script-compilacion-jar

  user@master$ cd 02-mapreduce
  user@master$ sh wc-gen-jar.sh

Para ejecutar:

  user@master$ hadoop jar wc.jar WordCount hdfs:///datasets/gutenberg-txt-es/*.txt hdfs:///user/<username>/data_out1

(puede tomar varios minutos)

  • el comando hadoop se este abandonando por yarn:
  user@master$ yarn jar wc.jar WordCount hdfs:///datasets/gutenberg-txt-es/*.txt hdfs:///user/<username>/data_out2

    //
    // WordCount.java
    //
    import java.io.IOException;
    import java.util.StringTokenizer;

    import org.apache.hadoop.conf.Configuration;
    import org.apache.hadoop.fs.Path;
    import org.apache.hadoop.io.IntWritable;
    import org.apache.hadoop.io.Text;
    import org.apache.hadoop.mapreduce.Job;
    import org.apache.hadoop.mapreduce.Mapper;
    import org.apache.hadoop.mapreduce.Reducer;
    import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
    import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;

    public class WordCount {

      public static class TokenizerMapper
           extends Mapper<Object, Text, Text, IntWritable>{

        private final static IntWritable one = new IntWritable(1);
        private Text word = new Text();

        public void map(Object key, Text value, Context context
                        ) throws IOException, InterruptedException {
          StringTokenizer itr = new StringTokenizer(value.toString());
          while (itr.hasMoreTokens()) {
            word.set(itr.nextToken());
            context.write(word, one);
          }
        }
      }

      public static class IntSumReducer
           extends Reducer<Text,IntWritable,Text,IntWritable> {
        private IntWritable result = new IntWritable();

        public void reduce(Text key, Iterable<IntWritable> values,
                           Context context
                           ) throws IOException, InterruptedException {
          int sum = 0;
          for (IntWritable val : values) {
            sum += val.get();
          }
          result.set(sum);
          context.write(key, result);
        }
      }

      public static void main(String[] args) throws Exception {
        Configuration conf = new Configuration();
        Job job = Job.getInstance(conf, "word count");
        job.setJarByClass(WordCount.class);
        job.setMapperClass(TokenizerMapper.class);
        job.setCombinerClass(IntSumReducer.class);
        job.setReducerClass(IntSumReducer.class);
        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(IntWritable.class);
        FileInputFormat.addInputPath(job, new Path(args[0]));
        FileOutputFormat.setOutputPath(job, new Path(args[1]));
        System.exit(job.waitForCompletion(true) ? 0 : 1);
      }
    }

(2) WordCount en python

  • Hay varias librerias de python para acceder a servicios MapReduce en Hadoop

  • Se usará MRJOB (https://pythonhosted.org/mrjob/)

  • Se puede emplear una version de python 2.x o 3.x, del sistema (como root) o con un manejador de versiones de node (pyenv o virtualenv).

  • Como parte del sistema, se instalará mrjob así:

  user@master$ sudo yum install python-pip
  user@master$ sudo pip install --upgrade pip
  user@master$ sudo pip install mrjob
  • Si utilizará un manejador de versiones de python, puede ser así:

primero instalar pyenv (https://github.com/pyenv/pyenv-installer)

  user@master$ curl -L https://raw.githubusercontent.com/pyenv/pyenv-installer/master/bin/pyenv-installer | bash
  user@master$ pyenv update
  user@master$ pyenv install 2.7.13
  user@master$ pyenv local 2.7.13
  user@master$ pip install mrjob
  • Probar mrjob python local:
  user@master$ cd 02-mapreduce
  user@master$ python wordcount-mr.py /datasets/gutenberg-txt-es/1*.txt
  • Ejecutar mrjob python en Hadoop con datos en hdfs:

Crear variable de ambiente: HADOOP_STREAMING_HOME para HORTONWORKS:

user@master$ export HADOOP_STREAMING_HOME=/usr/hdp/current/hadoop-mapreduce-client

Crear variable de ambiente: HADOOP_STREAMING_HOME para CLOUDERA CDH 5.14:

user@master$ export HADOOP_STREAMING_HOME=/opt/cloudera/parcels/CDH-5.14.0-1.cdh5.14.0.p0.24

Ejecutar:

  user@master$ python wordcount-mr.py hdfs:///datasets/gutenberg-txt-es/*.txt -r hadoop --output-dir hdfs:///user/<username>/data_out1 --hadoop-streaming-jar $HADOOP_STREAMING_HOME/hadoop-streaming.jar
  • el directorio 'data_out1' no puede existir)