public static void main(String[] args)
throws IOException, ClassNotFoundException, InterruptedException {
// TODO Auto-generated method stub
Configuration conf = new Configuration();
conf.set("mapreduce.input.keyvaluelinerecordreader.key.value.separator", ",");
Job job = Job.getInstance(conf);
job.setJarByClass(MapsideJoinDriver.class);
job.setMapperClass(MapsideJoinMapper.class);
job.setInputFormatClass(CompositeInputFormat.class);
String expr =
CompositeInputFormat.compose(
"inner", KeyValueTextInputFormat.class, new Path(args[0]), new Path(args[1]));
// String expr = CompositeInputFormat.compose("outer",KeyValueTextInputFormat.class , new
// Path(args[0]),new Path(args[1]));
job.getConfiguration().set("mapreduce.join.expr", expr);
job.setNumReduceTasks(0);
TextOutputFormat.setOutputPath(job, new Path(args[2]));
job.waitForCompletion(true);
}
public int run(
String oldRankVectorInput, String newRankVectorInput, String output, String l1NormOutput)
throws Exception {
Configuration conf = new Configuration();
conf.set("rightHandSideSum", rightHandSideSum + "");
conf.set("rankVectorSum", rankVectorSum + "");
Job job = new Job(conf, "Rank Vector Normalize");
job.setJarByClass(RankVectorNormalize.class);
job.setMapperClass(RankVectorNormalize.RankVectorNormalizeMapper.class);
job.setReducerClass(RankVectorNormalize.RankVectorNormalizeReducer.class);
job.setMapOutputKeyClass(LongWritable.class);
job.setMapOutputValueClass(BlockVectorElement.class);
job.setOutputKeyClass(LongWritable.class);
job.setOutputValueClass(BlockVectorElement.class);
job.setInputFormatClass(CompositeInputFormat.class);
String joinStatement =
CompositeInputFormat.compose(
"inner", SequenceFileInputFormat.class, oldRankVectorInput, newRankVectorInput);
conf.set("mapred.join.expr", joinStatement);
job.setOutputFormatClass(SequenceFileOutputFormat.class);
FileOutputFormat.setOutputPath(job, new Path(output));
MultipleOutputs.addNamedOutput(
job,
l1NormOutput,
SequenceFileOutputFormat.class,
NullWritable.class,
DoubleWritable.class);
return job.waitForCompletion(true) ? 0 : 1;
}
/**
* The main driver for sort program. Invoke this method to submit the map/reduce job.
*
* @throws Exception When there is communication problems with the job tracker.
*/
public int run(String[] args) throws Exception {
Configuration conf = getConf();
JobClient client = new JobClient(conf);
ClusterStatus cluster = client.getClusterStatus();
int num_reduces = (int) (cluster.getMaxReduceTasks() * 0.9);
String join_reduces = conf.get(REDUCES_PER_HOST);
if (join_reduces != null) {
num_reduces = cluster.getTaskTrackers() * Integer.parseInt(join_reduces);
}
Job job = new Job(conf);
job.setJobName("join");
job.setJarByClass(Sort.class);
job.setMapperClass(Mapper.class);
job.setReducerClass(Reducer.class);
Class<? extends InputFormat> inputFormatClass = SequenceFileInputFormat.class;
Class<? extends OutputFormat> outputFormatClass = SequenceFileOutputFormat.class;
Class<? extends WritableComparable> outputKeyClass = BytesWritable.class;
Class<? extends Writable> outputValueClass = TupleWritable.class;
String op = "inner";
List<String> otherArgs = new ArrayList<String>();
for (int i = 0; i < args.length; ++i) {
try {
if ("-r".equals(args[i])) {
num_reduces = Integer.parseInt(args[++i]);
} else if ("-inFormat".equals(args[i])) {
inputFormatClass = Class.forName(args[++i]).asSubclass(InputFormat.class);
} else if ("-outFormat".equals(args[i])) {
outputFormatClass = Class.forName(args[++i]).asSubclass(OutputFormat.class);
} else if ("-outKey".equals(args[i])) {
outputKeyClass = Class.forName(args[++i]).asSubclass(WritableComparable.class);
} else if ("-outValue".equals(args[i])) {
outputValueClass = Class.forName(args[++i]).asSubclass(Writable.class);
} else if ("-joinOp".equals(args[i])) {
op = args[++i];
} else {
otherArgs.add(args[i]);
}
} catch (NumberFormatException except) {
System.out.println("ERROR: Integer expected instead of " + args[i]);
return printUsage();
} catch (ArrayIndexOutOfBoundsException except) {
System.out.println("ERROR: Required parameter missing from " + args[i - 1]);
return printUsage(); // exits
}
}
// Set user-supplied (possibly default) job configs
job.setNumReduceTasks(num_reduces);
if (otherArgs.size() < 2) {
System.out.println("ERROR: Wrong number of parameters: ");
return printUsage();
}
FileOutputFormat.setOutputPath(job, new Path(otherArgs.remove(otherArgs.size() - 1)));
List<Path> plist = new ArrayList<Path>(otherArgs.size());
for (String s : otherArgs) {
plist.add(new Path(s));
}
job.setInputFormatClass(CompositeInputFormat.class);
job.getConfiguration()
.set(
CompositeInputFormat.JOIN_EXPR,
CompositeInputFormat.compose(op, inputFormatClass, plist.toArray(new Path[0])));
job.setOutputFormatClass(outputFormatClass);
job.setOutputKeyClass(outputKeyClass);
job.setOutputValueClass(outputValueClass);
Date startTime = new Date();
System.out.println("Job started: " + startTime);
int ret = job.waitForCompletion(true) ? 0 : 1;
Date end_time = new Date();
System.out.println("Job ended: " + end_time);
System.out.println(
"The job took " + (end_time.getTime() - startTime.getTime()) / 1000 + " seconds.");
return ret;
}