[20] | 1 | /** |
---|
| 2 | * Program: HBaseRecordPro.java |
---|
| 3 | * Editor: Waue Chen |
---|
| 4 | * From : NCHC. Taiwn |
---|
| 5 | * Last Update Date: 07/02/2008 |
---|
| 6 | * Upgrade to 0.17 |
---|
| 7 | */ |
---|
| 8 | /* |
---|
| 9 | * Cloud9: A MapReduce Library for Hadoop |
---|
| 10 | */ |
---|
| 11 | |
---|
| 12 | package tw.org.nchc.demo; |
---|
| 13 | |
---|
| 14 | import java.io.IOException; |
---|
| 15 | import java.util.Iterator; |
---|
| 16 | import java.util.StringTokenizer; |
---|
| 17 | |
---|
| 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.mapred.JobClient; |
---|
| 22 | import org.apache.hadoop.mapred.JobConf; |
---|
| 23 | import org.apache.hadoop.mapred.MapReduceBase; |
---|
| 24 | import org.apache.hadoop.mapred.Mapper; |
---|
| 25 | import org.apache.hadoop.mapred.OutputCollector; |
---|
| 26 | import org.apache.hadoop.mapred.Reducer; |
---|
| 27 | import org.apache.hadoop.mapred.Reporter; |
---|
| 28 | import org.apache.hadoop.mapred.SequenceFileInputFormat; |
---|
| 29 | import org.apache.hadoop.mapred.SequenceFileOutputFormat; |
---|
| 30 | |
---|
| 31 | import tw.org.nchc.code.Convert; |
---|
| 32 | import tw.org.nchc.tuple.Schema; |
---|
| 33 | import tw.org.nchc.tuple.Tuple; |
---|
| 34 | |
---|
| 35 | /** |
---|
| 36 | * <p> |
---|
| 37 | * Demo that illustrates the use of the tuple library ({@link Tuple} class). |
---|
| 38 | * Input comes from Bible+Shakespeare sample collection, encoded as single-field |
---|
| 39 | * tuples; see {@link DemoPackRecords}. Sample of final output: |
---|
| 40 | * </p> |
---|
| 41 | * |
---|
| 42 | * <pre> |
---|
| 43 | * ... |
---|
| 44 | * (admirable, 0) 9 |
---|
| 45 | * (admirable, 1) 6 |
---|
| 46 | * (admiral, 0) 2 |
---|
| 47 | * (admiral, 1) 4 |
---|
| 48 | * (admiration, 0) 10 |
---|
| 49 | * (admiration, 1) 6 |
---|
| 50 | * (admire, 0) 5 |
---|
| 51 | * (admire, 1) 3 |
---|
| 52 | * (admired, 0) 12 |
---|
| 53 | * (admired, 1) 7 |
---|
| 54 | * ... |
---|
| 55 | * </pre> |
---|
| 56 | * |
---|
| 57 | * <p> |
---|
| 58 | * The first field of the key tuple contains a token, the second field indicates |
---|
| 59 | * whether it was found on a even-length or odd-length line. The value is the |
---|
| 60 | * count of the tuple occurrences in the collection. In the MapReduce cycle, |
---|
| 61 | * output keys consist of tuples (Token, EvenOrOdd). The second field of the |
---|
| 62 | * tuple indicates whether the token was found on a line with an even or an odd |
---|
| 63 | * number of characters. Values consist of counts of tuple occurrences. Expected |
---|
| 64 | * trace of the demo: |
---|
| 65 | * </p> |
---|
| 66 | * |
---|
| 67 | * <pre> |
---|
| 68 | * Map input records=156215 |
---|
| 69 | * Map output records=1734298 |
---|
| 70 | * Map input bytes=13118917 |
---|
| 71 | * Map output bytes=66214039 |
---|
| 72 | * Combine input records=1734298 |
---|
| 73 | * Combine output records=192045 |
---|
| 74 | * Reduce input groups=59225 |
---|
| 75 | * Reduce input records=192045 |
---|
| 76 | * Reduce output records=59225 |
---|
| 77 | * </pre> |
---|
| 78 | * |
---|
| 79 | * <p> |
---|
| 80 | * Obviously, this isn't a particularly meaningful program, but does illustrate |
---|
| 81 | * the use of the {@link Tuple} class. |
---|
| 82 | * </p> |
---|
| 83 | */ |
---|
| 84 | public class DemoWordCountTuple { |
---|
| 85 | |
---|
| 86 | // create the schema for the tuple that will serve as the key |
---|
| 87 | private static final Schema KEY_SCHEMA = new Schema(); |
---|
| 88 | |
---|
| 89 | // define the schema statically |
---|
| 90 | static { |
---|
| 91 | KEY_SCHEMA.addField("Token", String.class, ""); |
---|
| 92 | KEY_SCHEMA.addField("EvenOrOdd", Integer.class, new Integer(1)); |
---|
| 93 | } |
---|
| 94 | |
---|
| 95 | // mapper that emits tuple as the key, and value '1' for each occurrence |
---|
| 96 | private static class MapClass extends MapReduceBase implements |
---|
| 97 | Mapper<LongWritable, Tuple, Tuple, IntWritable> { |
---|
| 98 | |
---|
| 99 | // define value '1' statically so we can reuse the object, i.e., avoid |
---|
| 100 | // unnecessary object creation |
---|
| 101 | private final static IntWritable one = new IntWritable(1); |
---|
| 102 | |
---|
| 103 | // once again, reuse tuples if possible |
---|
| 104 | private Tuple tupleOut = KEY_SCHEMA.instantiate(); |
---|
| 105 | |
---|
| 106 | public void map(LongWritable key, Tuple tupleIn, |
---|
| 107 | OutputCollector<Tuple, IntWritable> output, Reporter reporter) |
---|
| 108 | throws IOException { |
---|
| 109 | |
---|
| 110 | // the input value is a tuple; get field 0 |
---|
| 111 | // see DemoPackRecords of how input SequenceFile is generated |
---|
| 112 | String line = (String) tupleIn.get(0); |
---|
| 113 | StringTokenizer itr = new StringTokenizer(line); |
---|
| 114 | while (itr.hasMoreTokens()) { |
---|
| 115 | String token = itr.nextToken(); |
---|
| 116 | |
---|
| 117 | // put new values into the tuple |
---|
| 118 | tupleOut.set("Token", token); |
---|
| 119 | tupleOut.set("EvenOrOdd", line.length() % 2); |
---|
| 120 | |
---|
| 121 | // emit key-value pair |
---|
| 122 | output.collect(tupleOut, one); |
---|
| 123 | } |
---|
| 124 | } |
---|
| 125 | } |
---|
| 126 | |
---|
| 127 | // reducer counts up tuple occurrences |
---|
| 128 | private static class ReduceClass extends MapReduceBase implements |
---|
| 129 | Reducer<Tuple, IntWritable, Tuple, IntWritable> { |
---|
| 130 | private final static IntWritable SumValue = new IntWritable(); |
---|
| 131 | |
---|
| 132 | public synchronized void reduce(Tuple tupleKey, |
---|
| 133 | Iterator<IntWritable> values, |
---|
| 134 | OutputCollector<Tuple, IntWritable> output, Reporter reporter) |
---|
| 135 | throws IOException { |
---|
| 136 | // sum values |
---|
| 137 | int sum = 0; |
---|
| 138 | while (values.hasNext()) { |
---|
| 139 | sum += values.next().get(); |
---|
| 140 | } |
---|
| 141 | |
---|
| 142 | // keep original tuple key, emit sum of counts as value |
---|
| 143 | SumValue.set(sum); |
---|
| 144 | output.collect(tupleKey, SumValue); |
---|
| 145 | } |
---|
| 146 | } |
---|
| 147 | |
---|
| 148 | // dummy constructor |
---|
| 149 | private DemoWordCountTuple() { |
---|
| 150 | } |
---|
| 151 | |
---|
| 152 | /** |
---|
| 153 | * Runs the demo. |
---|
| 154 | */ |
---|
| 155 | public static void main(String[] args) throws IOException { |
---|
| 156 | String inPath = "/shared/sample-input/bible+shakes.nopunc.packed"; |
---|
| 157 | String outputPath = "word-counts-tuple"; |
---|
| 158 | int numMapTasks = 20; |
---|
| 159 | int numReduceTasks = 20; |
---|
| 160 | |
---|
| 161 | JobConf conf = new JobConf(DemoWordCountTuple.class); |
---|
| 162 | conf.setJobName("wordcount"); |
---|
| 163 | |
---|
| 164 | conf.setNumMapTasks(numMapTasks); |
---|
| 165 | conf.setNumReduceTasks(numReduceTasks); |
---|
| 166 | // 0.16 |
---|
| 167 | // conf.setInputPath(new Path(inPath)); |
---|
| 168 | Convert.setInputPath(conf, new Path(inPath)); |
---|
| 169 | conf.setInputFormat(SequenceFileInputFormat.class); |
---|
| 170 | |
---|
| 171 | // conf.setOutputPath(new Path(outputPath)); |
---|
| 172 | Convert.setInputPath(conf, new Path(outputPath)); |
---|
| 173 | conf.setOutputKeyClass(Tuple.class); |
---|
| 174 | conf.setOutputValueClass(IntWritable.class); |
---|
| 175 | conf.setOutputFormat(SequenceFileOutputFormat.class); |
---|
| 176 | |
---|
| 177 | conf.setMapperClass(MapClass.class); |
---|
| 178 | conf.setCombinerClass(ReduceClass.class); |
---|
| 179 | conf.setReducerClass(ReduceClass.class); |
---|
| 180 | |
---|
| 181 | JobClient.runJob(conf); |
---|
| 182 | } |
---|
| 183 | } |
---|