Example usage for org.apache.hadoop.mapred FileInputFormat setInputPaths

List of usage examples for org.apache.hadoop.mapred FileInputFormat setInputPaths

  1. HOME
  2. Java
  3. org.apache.hadoop
  4. org.apache.hadoop.mapred.*
  5. FileInputFormat
  6. setInputPaths

Introduction

In this page you can find the example usage for org.apache.hadoop.mapred FileInputFormat setInputPaths.

Prototype

public static void setInputPaths(JobConf conf, Path... inputPaths)

Source Link

Document

Set the array of Path s as the list of inputs for the map-reduce job.

Usage

From source file:hibench.DataGenerator.java

License:Apache License

public void sumUpZipf(Path fin, Path fout, ZipfRandom zipf) throws IOException {
    LOG.info("Summing up Zipfian Id Distirubtion...");

    JobConf job = new JobConf(WebDataGen.class);
    String jobname = fin.getName() + " -> " + fout.getName();
    job.setJobName(jobname);//  w w  w .  j  a  v  a 2  s  . c om

    zipf.setJobConf(job);

    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(Text.class);

    job.setMapperClass(SumUpZipfMapper.class);
    job.setReducerClass(SumUpZipfReducer.class);

    job.setNumReduceTasks(1); // Important to sequentially accumulate the required space

    job.setInputFormat(NLineInputFormat.class);
    job.setOutputFormat(TextOutputFormat.class);

    DataPaths.checkHdfsFile(fout, false);

    FileInputFormat.setInputPaths(job, fin);
    FileOutputFormat.setOutputPath(job, fout);

    LOG.info("Running Job: " + jobname);
    LOG.info("Dummy file: " + fin);
    LOG.info("Zipfian sum up file as Ouput: " + fout);
    JobClient.runJob(job);
    LOG.info("Finished Running Job: " + jobname);
}

From source file:hibench.DataGenerator.java

License:Apache License

public void createZipf(Path fin, Path fout, ZipfRandom zipf) throws IOException {
    LOG.info("Creating Zipfian Id Distirubtion...");

    JobConf job = new JobConf(WebDataGen.class);
    String jobname = fin.getName() + " -> " + fout.getName();

    job.setJobName(jobname);/*from  w  w w. j av a 2 s .c  o m*/

    zipf.setJobConf(job);

    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(Text.class);

    job.setMapperClass(CreateZipfDistrMapper.class);

    job.setNumReduceTasks(0);

    job.setInputFormat(NLineInputFormat.class);
    job.setOutputFormat(TextOutputFormat.class);

    FileInputFormat.setInputPaths(job, fin);
    FileOutputFormat.setOutputPath(job, fout);

    DataPaths.checkHdfsFile(fout, false);

    LOG.info("Running Job: " + jobname);
    LOG.info("Zipfian Sum File: " + fin);
    LOG.info("Zipfian Id distribution as Ouput: " + fout);
    RunningJob jobCreateZipf = JobClient.runJob(job);
    LOG.info("Finished Running Job: " + jobname);

    long vElems = jobCreateZipf.getCounters().getCounter(MAP_OUTPUT_RECORDS);
    LOG.info("Created " + vElems + " virtual zipfian elements");
    zipf.setVirtElems(vElems);
}

From source file:hibench.DataGenerator.java

License:Apache License

public void createHtmlPages(Path dummy, HtmlConf html) throws IOException {

    LOG.info("Creating Html Pages...");

    Path fout = new Path(dummy.getParent(), "tmp");

    JobConf job = new JobConf(WebDataGen.class);
    String jobname = "Create html pages to " + fout.getName();

    job.setJobName(jobname);/*from  w ww. j a  v a 2s.c o m*/

    html.setJobConf(job);

    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(Text.class);

    job.setMapperClass(CreateHtmlPagesMapper.class);

    job.setNumReduceTasks(0);

    FileInputFormat.setInputPaths(job, dummy);

    // first create result files under tmp folder
    FileOutputFormat.setOutputPath(job, fout);

    // begin from dummy file
    job.setInputFormat(NLineInputFormat.class);

    // use MultipleTextOutputFormat to produce three out files defined
    // in PathConf, i.e., LINK, PAGE_ZLINK_TABLE, PAGE_ZWORD_TABLE
    job.setOutputFormat(HtmlMultipleTextOutputFormat.class);

    LOG.info("Running Job: " + jobname);
    LOG.info("Dummy file: " + dummy);
    LOG.info("Multiple result Html files as <links, words, urls>");
    JobClient.runJob(job);
    LOG.info("Finished Running Job: " + jobname);

    // Move result files under tmp into parent path
    // and remove the empty tmp path finally 
    DataPaths.moveFilesToParent(fout);
}

From source file:hibench.PageRankDataGenerator.java

License:Apache License

private void createPageRankNodes() throws IOException {

    LOG.info("Creating PageRank nodes...", null);

    JobConf job = new JobConf(WebDataGen.class);
    String jobname = "Create " + paths.dname + " pagerank nodes";

    job.setJobName(jobname);/* w  w  w  . ja  v  a  2s  . c  om*/

    job.setOutputKeyClass(NullWritable.class);
    job.setOutputValueClass(Text.class);

    FileInputFormat.setInputPaths(job, paths.getPath(DataPaths.LINKS));
    job.setInputFormat(TextInputFormat.class);

    if (options.PAGERANK_NODE_BALANCE) {
        /***
         * Balance the output order of nodes, to prevent the running
         * of pagerank bench from potential data skew
         */
        job.setMapOutputKeyClass(LongWritable.class);
        job.setMapOutputValueClass(NullWritable.class);

        job.setMapperClass(BalancedLinkNodesMapper.class);
        job.setReducerClass(BalancedLinkNodesReducer.class);
        //         job.setPartitionerClass(ModulusPartitioner.class);

        if (options.reds > 0) {
            job.setNumReduceTasks(options.reds);
        } else {
            job.setNumReduceTasks(DataOptions.getMaxNumReduce());
        }
    } else {
        job.setMapOutputKeyClass(Text.class);

        job.setMapperClass(OutputLinkNodesMapper.class);
        job.setNumReduceTasks(0);
    }

    if (options.SEQUENCE_OUT) {
        job.setOutputFormat(SequenceFileOutputFormat.class);
    } else {
        job.setOutputFormat(TextOutputFormat.class);
    }

    if (null != options.codecClass) {
        job.set("mapred.output.compression.type", "BLOCK");
        FileOutputFormat.setCompressOutput(job, true);
        FileOutputFormat.setOutputCompressorClass(job, options.codecClass);
    }

    FileOutputFormat.setOutputPath(job, paths.getResult(DataPaths.VERTICALS));

    LOG.info("Running Job: " + jobname);
    LOG.info("Links file " + paths.getPath(DataPaths.LINKS) + " as input");
    LOG.info("Vertices file " + paths.getResult(DataPaths.VERTICALS) + " as output");
    JobClient.runJob(job);
    LOG.info("Finished Running Job: " + jobname);

    LOG.info("Cleaning temp files...");
    paths.cleanTempFiles(paths.getResult(DataPaths.VERTICALS));
}

From source file:hibench.PageRankDataGenerator.java

License:Apache License

/***
 * Create pagerank edge table, output link A->B as <A, B> pairs
 * @throws IOException//from ww w.  j  a  va2s  . com
 */
private void createPageRankLinks() throws IOException {

    LOG.info("Creating PageRank links", null);

    JobConf job = new JobConf(WebDataGen.class);
    String jobname = "Create " + paths.dname + " pagerank links";

    job.setJobName(jobname);

    job.setOutputKeyClass(NullWritable.class);
    job.setOutputValueClass(Text.class);
    job.setMapOutputKeyClass(Text.class);

    job.setNumReduceTasks(0);

    FileInputFormat.setInputPaths(job, paths.getPath(DataPaths.T_LINK_PAGE));
    job.setInputFormat(TextInputFormat.class);

    job.setMapperClass(OutputLinkEdgesMapper.class);

    if (options.SEQUENCE_OUT) {
        job.setOutputFormat(SequenceFileOutputFormat.class);
    } else {
        job.setOutputFormat(TextOutputFormat.class);
    }

    if (null != options.codecClass) {
        job.set("mapred.output.compression.type", "BLOCK");
        FileOutputFormat.setCompressOutput(job, true);
        FileOutputFormat.setOutputCompressorClass(job, options.codecClass);
    }

    FileOutputFormat.setOutputPath(job, paths.getResult(DataPaths.EDGES));

    LOG.info("Running Job: " + jobname);
    LOG.info("Table link-page " + paths.getPath(DataPaths.T_LINK_PAGE) + " as input");
    LOG.info("Edges file " + paths.getResult(DataPaths.EDGES) + " as output");
    JobClient.runJob(job);
    LOG.info("Finished Running Job: " + jobname);

    LOG.info("Cleaning temp files...");
    paths.cleanTempFiles(paths.getResult(DataPaths.EDGES));
}

From source file:hitune.analysis.mapreduce.processor.HadoopMetrics.java

License:Apache License

@Override
public void run() {
    // TODO Auto-generated method stub
    long timestamp = System.currentTimeMillis();
    JobConf conf = new JobConf(this.conf, HadoopMetrics.class);
    try {//from  w ww. ja  v  a2 s  .c  o  m
        conf.setJobName(this.getClass().getSimpleName() + timestamp);
        conf.setInputFormat(MultiSequenceFileInputFormat.class);
        conf.setMapperClass(HadoopMetrics.MapClass.class);
        conf.setReducerClass(SystemLog.ReduceClass.class);
        Class<? extends WritableComparable> outputKeyClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputKeyClass))
                .asSubclass(WritableComparable.class);
        Class<? extends Writable> outputValueClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputValueClass))
                .asSubclass(Writable.class);
        conf.setMapOutputKeyClass(outputKeyClass);
        conf.setMapOutputValueClass(outputValueClass);

        conf.setOutputKeyClass(Text.class);

        conf.setOutputValueClass(TextArrayWritable.class);
        conf.setOutputFormat(CSVFileOutputFormat.class);

        String outputPaths = conf.get(AnalysisProcessorConfiguration.reportfolder) + "/"
                + conf.get(AnalysisProcessorConfiguration.reportfile);
        String temp_outputPaths = getTempOutputDir(outputPaths);

        if (this.inputfiles != null) {
            log.debug("inputPaths:" + inputfiles);
            FileInputFormat.setInputPaths(conf, inputfiles);
            FileOutputFormat.setOutputPath(conf, new Path(temp_outputPaths));
            try {
                JobClient.runJob(conf);
                moveResults(conf, outputPaths, temp_outputPaths);
            } catch (IOException e) {
                // TODO Auto-generated catch block
                log.warn("For " + getOutputFileName() + " :JOB fails!");
                log.warn(e);
                e.printStackTrace();
                this.MOVE_DONE = false;
            }

        } else {
            log.warn("For " + getOutputFileName() + " :No input path!");
        }

    } catch (Exception e) {

        log.warn("Job preparation failure!");
        log.warn(e);
        e.printStackTrace();
    }

}

From source file:hitune.analysis.mapreduce.processor.HistoryLog.java

License:Apache License

public void run() {
    // TODO Auto-generated method stub
    long timestamp = System.currentTimeMillis();
    JobConf conf = new JobConf(this.conf, HistoryLog.class);
    try {//from   w  w  w.jav  a  2  s. com
        conf.setJobName(this.getClass().getSimpleName() + timestamp);
        conf.setInputFormat(MultiSequenceFileInputFormat.class);
        conf.setMapperClass(HistoryLog.MapClass.class);
        conf.setReducerClass(HistoryLog.ReduceClass.class);
        conf.setOutputKeyClass(Text.class);

        Class<? extends WritableComparable> outputKeyClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputKeyClass))
                .asSubclass(WritableComparable.class);
        Class<? extends Writable> outputValueClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputValueClass))
                .asSubclass(Writable.class);
        conf.setMapOutputKeyClass(outputKeyClass);
        conf.setMapOutputValueClass(outputValueClass);

        conf.setOutputValueClass(TextArrayWritable.class);
        conf.setOutputFormat(CSVFileOutputFormat.class);

        String outputPaths = conf.get(AnalysisProcessorConfiguration.reportfolder) + "/"
                + conf.get(AnalysisProcessorConfiguration.reportfile);
        String temp_outputPaths = getTempOutputDir(outputPaths);

        if (this.inputfiles != null) {
            log.debug("inputPaths:" + inputfiles);
            FileInputFormat.setInputPaths(conf, inputfiles);
            FileOutputFormat.setOutputPath(conf, new Path(temp_outputPaths));

            try {
                JobClient.runJob(conf);
                moveResults(conf, outputPaths, temp_outputPaths);
            } catch (IOException e) {
                // TODO Auto-generated catch block
                log.warn("For " + getOutputFileName() + " :JOB fails!");
                log.warn(e);
                e.printStackTrace();
                this.MOVE_DONE = false;
            }

        } else {
            log.warn("For " + getOutputFileName() + " :No input path!");
        }
    } catch (Exception e) {
        log.warn("Job preparation failure!");
        log.warn(e);
        e.printStackTrace();
    }
}

From source file:hitune.analysis.mapreduce.processor.InstrumentDataflow.java

License:Apache License

@Override
public void run() {
    // TODO Auto-generated method stub

    long timestamp = System.currentTimeMillis();

    JobConf conf = new JobConf(this.conf, InstrumentDataflow.class);
    try {//w  w w.j  ava2 s.  c  o m
        conf.setJobName(this.getClass().getSimpleName() + timestamp);
        conf.setInputFormat(MultiSequenceFileInputFormat.class);
        conf.setMapperClass(InstrumentDataflow.MapClass.class);
        conf.setReducerClass(InstrumentDataflow.ReduceClass.class);
        conf.setOutputKeyClass(Text.class);
        Class<? extends WritableComparable> outputKeyClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputKeyClass))
                .asSubclass(WritableComparable.class);
        Class<? extends Writable> outputValueClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputValueClass))
                .asSubclass(Writable.class);
        conf.setMapOutputKeyClass(outputKeyClass);
        conf.setMapOutputValueClass(outputValueClass);

        conf.setOutputValueClass(TextArrayWritable.class);
        conf.setOutputFormat(CSVFileOutputFormat.class);

        String outputPaths = conf.get(AnalysisProcessorConfiguration.reportfolder) + "/"
                + conf.get(AnalysisProcessorConfiguration.reportfile);
        String temp_outputPaths = getTempOutputDir(outputPaths);

        if (this.inputfiles != null) {
            log.debug("inputPaths:" + inputfiles);
            FileInputFormat.setInputPaths(conf, inputfiles);
            FileOutputFormat.setOutputPath(conf, new Path(temp_outputPaths));

            //FileInputFormat.setInputPathFilter(conf, evtFileFilter.class);
            //conf.setNumReduceTasks(1);

            try {
                JobClient.runJob(conf);
                moveResults(conf, outputPaths, temp_outputPaths);
            } catch (IOException e) {
                // TODO Auto-generated catch block
                log.warn("For " + getOutputFileName() + " :JOB fails!");
                log.warn(e);
                e.printStackTrace();
                this.MOVE_DONE = false;
            }

        } else {
            log.warn("For " + getOutputFileName() + " :No input path!");
        }
    } catch (Exception e) {
        log.warn("Job preparation failure!");
        log.warn(e);
        e.printStackTrace();
    }
}

From source file:hitune.analysis.mapreduce.processor.InstrumentSamplingTop.java

License:Apache License

@Override
public void run() {
    // TODO Auto-generated method stub
    long timestamp = System.currentTimeMillis();
    try {//from  w  w w  . j  a  v  a 2s .  c om
        JobConf conf = new JobConf(this.conf, InstrumentSamplingTop.class);
        conf.setJobName(this.getClass().getSimpleName() + "_1_" + timestamp);

        conf.setInputFormat(MultiSequenceFileInputFormat.class);
        conf.setMapperClass(InstrumentSamplingTop.MapClass.class);
        conf.setReducerClass(InstrumentSamplingTop.ReduceClass.class);

        Class<? extends WritableComparable> outputKeyClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputKeyClass))
                .asSubclass(WritableComparable.class);
        Class<? extends Writable> outputValueClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputValueClass))
                .asSubclass(Writable.class);
        conf.setMapOutputKeyClass(outputKeyClass);
        conf.setMapOutputValueClass(outputValueClass);
        conf.setOutputKeyClass(outputKeyClass);
        conf.setOutputValueClass(outputValueClass);

        conf.setOutputFormat(SequenceFileOutputFormat.class);

        String outputPaths = conf.get(AnalysisProcessorConfiguration.reportfolder) + "/"
                + conf.get(AnalysisProcessorConfiguration.reportfile);

        String temp_outputPaths = getTempOutputDir(outputPaths);

        if (this.inputfiles != null) {
            log.debug("inputPaths:" + inputfiles);
            FileInputFormat.setInputPaths(conf, inputfiles);
            FileOutputFormat.setOutputPath(conf, new Path(outputPaths + "_1_" + timestamp));

            try {

                //first job
                JobClient.runJob(conf);

                JobConf secondconf = new JobConf(this.conf, InstrumentSamplingTop.class);
                secondconf.setJobName(this.getClass().getSimpleName() + "_2_" + timestamp);
                secondconf.setInputFormat(SequenceFileInputFormat.class);
                secondconf.setMapperClass(IdentityMapper.class);
                secondconf.setReducerClass(InstrumentSamplingTop.TopClass.class);

                secondconf.setMapOutputKeyClass(outputKeyClass);
                secondconf.setMapOutputValueClass(outputValueClass);

                secondconf.setOutputKeyClass(Text.class);
                secondconf.setOutputValueClass(TextArrayWritable.class);
                secondconf.setOutputFormat(CSVFileOutputFormat.class);
                FileInputFormat.setInputPaths(secondconf, outputPaths + "_1_" + timestamp);
                FileOutputFormat.setOutputPath(secondconf, new Path(temp_outputPaths));

                //second job to get ranking list
                JobClient.runJob(secondconf);
                moveResults(secondconf, outputPaths, temp_outputPaths);
                Path temp = new Path(outputPaths + "_1_" + timestamp);
                temp.getFileSystem(conf).delete(temp);
            } catch (IOException e) {
                // TODO Auto-generated catch block
                log.warn("For " + getOutputFileName() + " :JOB fails!");
                log.warn(e);
                e.printStackTrace();
                this.MOVE_DONE = false;
            }

        } else {
            log.warn("For " + getOutputFileName() + " :No input path!");
        }
    } catch (Exception e) {
        log.warn("Job preparation failure!");
        log.warn(e);
        e.printStackTrace();
    }

}

From source file:hitune.analysis.mapreduce.processor.SystemLog.java

License:Apache License

@Override
public void run() {

    // TODO Auto-generated method stub
    long timestamp = System.currentTimeMillis();
    JobConf conf = new JobConf(this.conf, SystemLog.class);

    try {//  w  w  w.j av  a 2 s .co m
        conf.setJobName(this.getClass().getSimpleName() + timestamp);

        conf.setInputFormat(MultiSequenceFileInputFormat.class);
        conf.setMapperClass(SystemLog.MapClass.class);
        conf.setReducerClass(SystemLog.ReduceClass.class);

        Class<? extends WritableComparable> outputKeyClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputKeyClass))
                .asSubclass(WritableComparable.class);
        Class<? extends Writable> outputValueClass = Class
                .forName(conf.get(AnalysisProcessorConfiguration.mapoutputValueClass))
                .asSubclass(Writable.class);
        conf.setMapOutputKeyClass(outputKeyClass);
        conf.setMapOutputValueClass(outputValueClass);

        conf.setOutputKeyClass(Text.class);

        conf.setOutputValueClass(TextArrayWritable.class);
        conf.setOutputFormat(CSVFileOutputFormat.class);

        String outputPaths = conf.get(AnalysisProcessorConfiguration.reportfolder) + "/"
                + conf.get(AnalysisProcessorConfiguration.reportfile);
        String temp_outputPaths = getTempOutputDir(outputPaths);

        if (this.inputfiles != null) {
            log.debug("inputPaths:" + inputfiles);
            FileInputFormat.setInputPaths(conf, inputfiles);
            FileOutputFormat.setOutputPath(conf, new Path(temp_outputPaths));
            try {
                JobClient.runJob(conf);
                moveResults(conf, outputPaths, temp_outputPaths);
            } catch (IOException e) {
                // TODO Auto-generated catch block
                log.warn("For " + getOutputFileName() + " :JOB fails!");
                log.warn(e);
                e.printStackTrace();
                this.MOVE_DONE = false;
            }

        } else {
            log.warn("For " + getOutputFileName() + " :No input path!");

        }

    } catch (Exception e) {
        log.warn("Job preparation failure!");
        log.warn(e);
        e.printStackTrace();
    }

}