List of usage examples for org.apache.hadoop.mapreduce Job setMapperClass
public void setMapperClass(Class<? extends Mapper> cls) throws IllegalStateException
From source file:com.bizosys.oneline.maintenance.Import.java
License:Apache License
/** * Sets up the actual job./*from w ww . j av a2 s . c o m*/ * * @param conf The current configuration. * @param args The command line parameters. * @return The newly created job. * @throws IOException When setting up the job fails. */ public static Job createSubmittableJob(Configuration conf, String[] args) throws IOException { String tableName = args[0]; Path inputDir = new Path(args[1]); Job job = new Job(conf, NAME + "_" + tableName); job.setJarByClass(Importer.class); System.out.println("Input Dir:" + inputDir); FileInputFormat.setInputPaths(job, inputDir); job.setInputFormatClass(SequenceFileInputFormat.class); job.setMapperClass(Importer.class); // No reducers. Just write straight to table. Call initTableReducerJob // because it sets up the TableOutputFormat. TableMapReduceUtil.initTableReducerJob(tableName, null, job); job.setNumReduceTasks(0); return job; }
From source file:com.blackberry.logdriver.util.Cat.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }/*from w w w . j av a2s. c o m*/ FileSystem fs = FileSystem.get(conf); // The command line options List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 2) { System.out.println("usage: [genericOptions] input [input ...] output"); System.exit(1); } // Get the files we need from the command line. for (int i = 0; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(Cat.class); jobConf.setIfUnset("mapred.job.name", "Cat Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } job.setInputFormatClass(BoomInputFormat.class); job.setMapperClass(CatMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { BoomInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.FastSearch.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }//from w ww . j a v a2 s . co m FileSystem fs = FileSystem.get(conf); // The command line options String searchString = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchString input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchString = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(FastSearch.class); jobConf.setIfUnset("mapred.job.name", "Search Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string", Base64.encodeBase64String(searchString.getBytes("UTF-8"))); job.setInputFormatClass(AvroBlockInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { AvroBlockInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.Grep.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }/*w w w . jav a 2 s. co m*/ FileSystem fs = FileSystem.get(conf); // The command line options String regex = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] regex input [input ...] output"); System.exit(1); } // Get the files we need from the command line. regex = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(Grep.class); jobConf.setIfUnset("mapred.job.name", "Grep Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.grep.regex", Base64.encodeBase64String(regex.getBytes("UTF-8"))); job.setInputFormatClass(BoomInputFormat.class); job.setMapperClass(GrepMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { BoomInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.MultiSearch.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }// w w w . j av a2s.c o m FileSystem fs = FileSystem.get(conf); // The command line options String searchStringDir = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchStringDirectory input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchStringDir = args[0]; // We are going to be reading all the files in this directory a lot. So // let's up the replication factor by a lot so that they're easy to read. for (FileStatus f : fs.listStatus(new Path(searchStringDir))) { fs.setReplication(f.getPath(), (short) 16); } for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(MultiSearch.class); jobConf.setIfUnset("mapred.job.name", "MultiSearch"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string.dir", searchStringDir); // This search is generally too fast to make good use of 128MB blocks, so // let's set the value to 256MB (if it's not set already) if (jobConf.get("mapred.max.split.size") == null) { jobConf.setLong("mapred.max.split.size", 256 * 1024 * 1024); } job.setInputFormatClass(AvroBlockInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { AvroBlockInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blackberry.logdriver.util.Search.java
License:Apache License
@Override public int run(String[] args) throws Exception { Configuration conf = getConf(); // Configuration processed by ToolRunner // If run by Oozie, then load the Oozie conf too if (System.getProperty("oozie.action.conf.xml") != null) { conf.addResource(new URL("file://" + System.getProperty("oozie.action.conf.xml"))); }/* www. ja v a 2 s.c om*/ FileSystem fs = FileSystem.get(conf); // The command line options String searchString = null; List<Path> paths = new ArrayList<Path>(); Path outputDir = null; // Load input files from the command line if (args.length < 3) { System.out.println("usage: [genericOptions] searchString input [input ...] output"); System.exit(1); } // Get the files we need from the command line. searchString = args[0]; for (int i = 1; i < args.length - 1; i++) { for (FileStatus f : fs.globStatus(new Path(args[i]))) { paths.add(f.getPath()); } } outputDir = new Path(args[args.length - 1]); @SuppressWarnings("deprecation") Job job = new Job(conf); Configuration jobConf = job.getConfiguration(); job.setJarByClass(Search.class); jobConf.setIfUnset("mapred.job.name", "Search Files"); // To propagate credentials within Oozie if (System.getenv("HADOOP_TOKEN_FILE_LOCATION") != null) { jobConf.set("mapreduce.job.credentials.binary", System.getenv("HADOOP_TOKEN_FILE_LOCATION")); } // Good output separators include things that are unsupported by XML. So we // just send the byte value of the character through. The restriction here // is that it can't be more than 1 byte when UTF-8 encoded, since it will be // read by Pig which only deals with single byte separators. { String outputSeparator = jobConf.get("logdriver.output.field.separator", DEFAULT_OUTPUT_SEPARATOR); byte[] bytes = outputSeparator.getBytes(UTF_8); if (bytes.length != 1) { LOG.error("The output separator must be a single byte in UTF-8."); return 1; } jobConf.set("logdriver.output.field.separator", Byte.toString(bytes[0])); } jobConf.set("logdriver.search.string", searchString); job.setInputFormatClass(BoomInputFormat.class); job.setMapperClass(SearchMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(NullWritable.class); job.setNumReduceTasks(0); // And set the output as usual job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, outputDir); for (Path path : paths) { BoomInputFormat.addInputPath(job, path); } // Run the job. if (conf.getBoolean("job.wait", DEFAULT_WAIT_JOB)) { return job.waitForCompletion(true) ? 0 : 1; } else { job.submit(); return 0; } }
From source file:com.blogclustermr.EdgeLister.java
public static void main(String[] args) throws Exception { Configuration conf = new Configuration(); Job job; job = Job.getInstance(conf, "edge_lister"); job.setJarByClass(EdgeLister.class); job.setMapperClass(EdgeMapper.class); //job.setCombinerClass(EdgeWeightReducer.class); job.setReducerClass(EdgeWeightReducer.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(LongWritable.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(LongWritable.class); FileInputFormat.addInputPath(job, new Path(args[0])); FileOutputFormat.setOutputPath(job, new Path(args[1])); job.waitForCompletion(true);/*from w w w .j a va 2 s . co m*/ System.exit(job.waitForCompletion(true) ? 0 : 1); }
From source file:com.cg.mapreduce.fpgrowth.mahout.fpm.PFPGrowth.java
License:Apache License
/** * Run the aggregation Job to aggregate the different TopK patterns and group each Pattern by the features * present in it and thus calculate the final Top K frequent Patterns for each feature *//*from ww w . ja v a 2 s.c o m*/ public static void startAggregating(Parameters params, Configuration conf) throws IOException, InterruptedException, ClassNotFoundException { conf.set(PFP_PARAMETERS, params.toString()); conf.set("mapred.compress.map.output", "true"); conf.set("mapred.output.compression.type", "BLOCK"); Path input = new Path(params.get(OUTPUT), FPGROWTH); Job job = new Job(conf, "PFP Aggregator Driver running over input: " + input); job.setJarByClass(PFPGrowth.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(TopKStringPatterns.class); FileInputFormat.addInputPath(job, input); Path outPath = new Path(params.get(OUTPUT), FREQUENT_PATTERNS); FileOutputFormat.setOutputPath(job, outPath); job.setInputFormatClass(SequenceFileInputFormat.class); job.setMapperClass(AggregatorMapper.class); job.setCombinerClass(AggregatorReducer.class); job.setReducerClass(AggregatorReducer.class); job.setOutputFormatClass(SequenceFileOutputFormat.class); HadoopUtil.delete(conf, outPath); boolean succeeded = job.waitForCompletion(true); if (!succeeded) { throw new IllegalStateException("Job failed!"); } }
From source file:com.cg.mapreduce.fpgrowth.mahout.fpm.PFPGrowth.java
License:Apache License
/** * Count the frequencies of various features in parallel using Map/Reduce *//*from w ww .j av a 2 s . c o m*/ public static void startParallelCounting(Parameters params, Configuration conf) throws IOException, InterruptedException, ClassNotFoundException { conf.set(PFP_PARAMETERS, params.toString()); conf.set("mapred.compress.map.output", "true"); conf.set("mapred.output.compression.type", "BLOCK"); String input = params.get(INPUT); Job job = new Job(conf, "Parallel Counting Driver running over input: " + input); job.setJarByClass(PFPGrowth.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(LongWritable.class); FileInputFormat.addInputPath(job, new Path(input)); Path outPath = new Path(params.get(OUTPUT), PARALLEL_COUNTING); FileOutputFormat.setOutputPath(job, outPath); HadoopUtil.delete(conf, outPath); job.setInputFormatClass(TextInputFormat.class); job.setMapperClass(ParallelCountingMapper.class); job.setCombinerClass(ParallelCountingReducer.class); job.setReducerClass(ParallelCountingReducer.class); job.setOutputFormatClass(SequenceFileOutputFormat.class); boolean succeeded = job.waitForCompletion(true); if (!succeeded) { throw new IllegalStateException("Job failed!"); } }
From source file:com.cg.mapreduce.fpgrowth.mahout.fpm.PFPGrowth.java
License:Apache License
/** * Run the Parallel FPGrowth Map/Reduce Job to calculate the Top K features of group dependent shards */// w w w.j a va 2 s . c om public static void startParallelFPGrowth(Parameters params, Configuration conf) throws IOException, InterruptedException, ClassNotFoundException { conf.set(PFP_PARAMETERS, params.toString()); conf.set("mapred.compress.map.output", "true"); conf.set("mapred.output.compression.type", "BLOCK"); Path input = new Path(params.get(INPUT)); Job job = new Job(conf, "PFP Growth Driver running over input" + input); job.setJarByClass(PFPGrowth.class); job.setMapOutputKeyClass(IntWritable.class); job.setMapOutputValueClass(TransactionTree.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(TopKStringPatterns.class); FileInputFormat.addInputPath(job, input); Path outPath = new Path(params.get(OUTPUT), FPGROWTH); FileOutputFormat.setOutputPath(job, outPath); HadoopUtil.delete(conf, outPath); job.setInputFormatClass(TextInputFormat.class); job.setMapperClass(ParallelFPGrowthMapper.class); job.setCombinerClass(ParallelFPGrowthCombiner.class); job.setReducerClass(ParallelFPGrowthReducer.class); job.setOutputFormatClass(SequenceFileOutputFormat.class); boolean succeeded = job.waitForCompletion(true); if (!succeeded) { throw new IllegalStateException("Job failed!"); } }