[21] | 1 | /** |
---|
| 2 | * Program: HBaseRecordPro.java |
---|
| 3 | * Editor: Waue Chen |
---|
| 4 | * From : NCHC. Taiwn |
---|
| 5 | * Last Update Date: 07/02/2008 |
---|
| 6 | */ |
---|
| 7 | /* |
---|
| 8 | * Cloud9: A MapReduce Library for Hadoop |
---|
| 9 | */ |
---|
| 10 | |
---|
| 11 | package tw.org.nchc.demo; |
---|
| 12 | |
---|
| 13 | import java.io.IOException; |
---|
| 14 | import java.util.Iterator; |
---|
| 15 | import java.util.StringTokenizer; |
---|
| 16 | |
---|
| 17 | import org.apache.hadoop.fs.FileSystem; |
---|
| 18 | import org.apache.hadoop.fs.Path; |
---|
| 19 | import org.apache.hadoop.io.IntWritable; |
---|
| 20 | import org.apache.hadoop.io.LongWritable; |
---|
| 21 | import org.apache.hadoop.io.Text; |
---|
| 22 | import org.apache.hadoop.mapred.JobClient; |
---|
| 23 | import org.apache.hadoop.mapred.JobConf; |
---|
| 24 | import org.apache.hadoop.mapred.MapReduceBase; |
---|
| 25 | import org.apache.hadoop.mapred.Mapper; |
---|
| 26 | import org.apache.hadoop.mapred.OutputCollector; |
---|
| 27 | import org.apache.hadoop.mapred.Reducer; |
---|
| 28 | import org.apache.hadoop.mapred.Reporter; |
---|
| 29 | |
---|
| 30 | /** |
---|
| 31 | * <p>Simple word count demo. Counts words in the Bible+Shakespeare sample |
---|
| 32 | * collection. Expected trace of MapReduce operation:</p> |
---|
| 33 | * |
---|
| 34 | * <pre> |
---|
| 35 | * Map input records=156215 |
---|
| 36 | * Map output records=1734298 |
---|
| 37 | * Map input bytes=9068074 |
---|
| 38 | * Map output bytes=15919397 |
---|
| 39 | * Combine input records=1734298 |
---|
| 40 | * Combine output records=135372 |
---|
| 41 | * Reduce input groups=41788 |
---|
| 42 | * Reduce input records=135372 |
---|
| 43 | * Reduce output records=41788 |
---|
| 44 | * </pre> |
---|
| 45 | * |
---|
| 46 | */ |
---|
| 47 | public class DemoWordCount { |
---|
| 48 | |
---|
| 49 | // mapper: emits (token, 1) for every word occurrence |
---|
| 50 | private static class MapClass extends MapReduceBase implements |
---|
| 51 | Mapper<LongWritable, Text, Text, IntWritable> { |
---|
| 52 | |
---|
| 53 | // reuse objects to save overhead of object creation |
---|
| 54 | private final static IntWritable one = new IntWritable(1); |
---|
| 55 | private Text word = new Text(); |
---|
| 56 | |
---|
| 57 | public void map(LongWritable key, Text value, |
---|
| 58 | OutputCollector<Text, IntWritable> output, Reporter reporter) |
---|
| 59 | throws IOException { |
---|
| 60 | String line = ((Text) value).toString(); |
---|
| 61 | StringTokenizer itr = new StringTokenizer(line); |
---|
| 62 | while (itr.hasMoreTokens()) { |
---|
| 63 | word.set(itr.nextToken()); |
---|
| 64 | output.collect(word, one); |
---|
| 65 | } |
---|
| 66 | } |
---|
| 67 | } |
---|
| 68 | |
---|
| 69 | // reducer: sums up all the counts |
---|
| 70 | private static class ReduceClass extends MapReduceBase implements |
---|
| 71 | Reducer<Text, IntWritable, Text, IntWritable> { |
---|
| 72 | |
---|
| 73 | // reuse objects |
---|
| 74 | private final static IntWritable SumValue = new IntWritable(); |
---|
| 75 | |
---|
| 76 | public void reduce(Text key, Iterator<IntWritable> values, |
---|
| 77 | OutputCollector<Text, IntWritable> output, Reporter reporter) |
---|
| 78 | throws IOException { |
---|
| 79 | // sum up values |
---|
| 80 | int sum = 0; |
---|
| 81 | while (values.hasNext()) { |
---|
| 82 | sum += values.next().get(); |
---|
| 83 | } |
---|
| 84 | SumValue.set(sum); |
---|
| 85 | output.collect(key, SumValue); |
---|
| 86 | } |
---|
| 87 | } |
---|
| 88 | |
---|
| 89 | private DemoWordCount() { |
---|
| 90 | } |
---|
| 91 | |
---|
| 92 | /** |
---|
| 93 | * Runs the demo. |
---|
| 94 | */ |
---|
| 95 | public static void main(String[] args) throws IOException { |
---|
| 96 | String filename = "/user/waue/test/132.txt"; |
---|
| 97 | String outputPath = "sample-counts"; |
---|
| 98 | int mapTasks = 20; |
---|
| 99 | int reduceTasks = 1; |
---|
| 100 | |
---|
| 101 | JobConf conf = new JobConf(DemoWordCount.class); |
---|
| 102 | conf.setJobName("wordcount"); |
---|
| 103 | |
---|
| 104 | conf.setNumMapTasks(mapTasks); |
---|
| 105 | conf.setNumReduceTasks(reduceTasks); |
---|
[25] | 106 | |
---|
| 107 | conf.setInputPath(new Path(filename)); |
---|
| 108 | |
---|
[21] | 109 | conf.setOutputKeyClass(Text.class); |
---|
| 110 | conf.setOutputValueClass(IntWritable.class); |
---|
[25] | 111 | |
---|
| 112 | conf.setOutputPath(new Path(outputPath)); |
---|
[21] | 113 | conf.setMapperClass(MapClass.class); |
---|
| 114 | conf.setCombinerClass(ReduceClass.class); |
---|
| 115 | conf.setReducerClass(ReduceClass.class); |
---|
| 116 | |
---|
| 117 | // Delete the output directory if it exists already |
---|
| 118 | Path outputDir = new Path(outputPath); |
---|
[25] | 119 | FileSystem.get(conf).delete(outputDir); |
---|
[21] | 120 | JobClient.runJob(conf); |
---|
| 121 | } |
---|
| 122 | } |
---|