List of usage examples for org.apache.hadoop.mapreduce Job setOutputKeyClass
public void setOutputKeyClass(Class<?> theClass) throws IllegalStateException
From source file:com.igalia.wordcount.WordCount.java
License:Open Source License
public int run(String[] arg0) throws Exception { Job job = new Job(getConf()); job.setJarByClass(WordCount.class); job.setJobName("wordcount"); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); job.setMapperClass(MapClass.class); job.setReducerClass(Reduce.class); FileInputFormat.setInputPaths(job, new Path("/tmp/wordcount/in")); FileOutputFormat.setOutputPath(job, new Path("/tmp/wordcount/out")); boolean success = job.waitForCompletion(true); return success ? 0 : 1; }
From source file:com.ikanow.aleph2.analytics.hadoop.assets.VerySimpleLocalExample.java
License:Apache License
@SuppressWarnings({ "deprecation", "unchecked", "rawtypes" }) @Test/* www .j av a2 s. c o m*/ public void test_localHadoopLaunch() throws IOException, IllegalStateException, ClassNotFoundException, InterruptedException { // 0) Setup the temp dir final String temp_dir = System.getProperty("java.io.tmpdir") + File.separator; //final Path tmp_path = FileContext.getLocalFSFileContext().makeQualified(new Path(temp_dir)); final Path tmp_path2 = FileContext.getLocalFSFileContext() .makeQualified(new Path(temp_dir + "/tmp_output")); try { FileContext.getLocalFSFileContext().delete(tmp_path2, true); } catch (Exception e) { } // (just doesn't exist yet) // 1) Setup config with local mode final Configuration config = new Configuration(); config.setBoolean("mapred.used.genericoptionsparser", true); // (just stops an annoying warning from appearing) config.set("fs.file.impl", "org.apache.hadoop.fs.LocalFileSystem"); config.set("mapred.job.tracker", "local"); config.set("fs.defaultFS", "local"); config.unset("mapreduce.framework.name"); // If running locally, turn "snappy" off - tomcat isn't pointing its native library path in the right place config.set("mapred.map.output.compression.codec", "org.apache.hadoop.io.compress.DefaultCodec"); // 2) Build job and do more setup using the Job API //TODO: not sure why this is deprecated, it doesn't seem to be in v1? We do need to move to JobConf at some point, but I ran into some // issues when trying to do everything I needed to for V1, so seems expedient to start here and migrate away final Job hj = new Job(config); // (NOTE: from here, changes to config are ignored) // Input format: //TOOD: fails because of guava issue, looks like we'll need to move to 2.7 and check it works with 2.5.x server? //TextInputFormat.addInputPath(hj, tmp_path); //hj.setInputFormatClass((Class<? extends InputFormat>) Class.forName ("org.apache.hadoop.mapreduce.lib.input.TextInputFormat")); hj.setInputFormatClass(TestInputFormat.class); // Output format: hj.setOutputFormatClass((Class<? extends OutputFormat>) Class .forName("org.apache.hadoop.mapreduce.lib.output.TextOutputFormat")); TextOutputFormat.setOutputPath(hj, tmp_path2); // Mapper etc (combiner/reducer are similar) hj.setMapperClass(TestMapper.class); hj.setOutputKeyClass(Text.class); hj.setOutputValueClass(Text.class); hj.setNumReduceTasks(0); // (disable reducer for now) hj.setJar("test"); try { hj.submit(); } catch (UnsatisfiedLinkError e) { throw new RuntimeException( "This is a windows/hadoop compatibility problem - adding the hadoop-commons in the misc_test_assets subdirectory to the top of the classpath should resolve it (and does in V1), though I haven't yet made that work with Aleph2", e); } //hj.getJobID().toString(); while (!hj.isComplete()) { Thread.sleep(1000); } assertTrue("Finished successfully", hj.isSuccessful()); }
From source file:com.ikanow.infinit.e.core.mapreduce.HadoopJobRunner.java
License:Open Source License
@SuppressWarnings({ "unchecked", "rawtypes" }) private String runHadoopJob(CustomMapReduceJobPojo job, String tempJarLocation) throws IOException, SAXException, ParserConfigurationException { StringWriter xml = new StringWriter(); createConfigXML(xml, job.jobtitle, job.inputCollection, getQueryOrProcessing(job.query, QuerySpec.INPUTFIELDS), job.isCustomTable, job.getOutputDatabase(), job._id.toString(), job.outputCollectionTemp, job.mapper, job.reducer, job.combiner, getQueryOrProcessing(job.query, QuerySpec.QUERY), job.communityIds, job.outputKey, job.outputValue, job.arguments);/*www . j a va 2s . c o m*/ ClassLoader savedClassLoader = Thread.currentThread().getContextClassLoader(); URLClassLoader child = new URLClassLoader(new URL[] { new File(tempJarLocation).toURI().toURL() }, savedClassLoader); Thread.currentThread().setContextClassLoader(child); // Now load the XML into a configuration object: Configuration config = new Configuration(); try { DocumentBuilderFactory dbFactory = DocumentBuilderFactory.newInstance(); DocumentBuilder dBuilder = dbFactory.newDocumentBuilder(); Document doc = dBuilder.parse(new ByteArrayInputStream(xml.toString().getBytes())); NodeList nList = doc.getElementsByTagName("property"); for (int temp = 0; temp < nList.getLength(); temp++) { Node nNode = nList.item(temp); if (nNode.getNodeType() == Node.ELEMENT_NODE) { Element eElement = (Element) nNode; String name = getTagValue("name", eElement); String value = getTagValue("value", eElement); if ((null != name) && (null != value)) { config.set(name, value); } } } } catch (Exception e) { throw new IOException(e.getMessage()); } // Now run the JAR file try { config.setBoolean("mapred.used.genericoptionsparser", true); // (just stops an annoying warning from appearing) if (bLocalMode) { config.set("mapred.job.tracker", "local"); config.set("fs.default.name", "local"); } else { String trackerUrl = HadoopUtils.getXMLProperty( prop_custom.getHadoopConfigPath() + "/hadoop/mapred-site.xml", "mapred.job.tracker"); String fsUrl = HadoopUtils.getXMLProperty( prop_custom.getHadoopConfigPath() + "/hadoop/core-site.xml", "fs.default.name"); config.set("mapred.job.tracker", trackerUrl); config.set("fs.default.name", fsUrl); } Job hj = new Job(config); Class<?> classToLoad = Class.forName(job.mapper, true, child); hj.setJarByClass(classToLoad); hj.setInputFormatClass((Class<? extends InputFormat>) Class .forName("com.ikanow.infinit.e.data_model.custom.InfiniteMongoInputFormat", true, child)); if ((null != job.exportToHdfs) && job.exportToHdfs) { hj.setOutputFormatClass((Class<? extends OutputFormat>) Class .forName("org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat", true, child)); Path outPath = this.ensureOutputDirectory(job); SequenceFileOutputFormat.setOutputPath(hj, outPath); } else { // normal case, stays in MongoDB hj.setOutputFormatClass((Class<? extends OutputFormat>) Class .forName("com.mongodb.hadoop.MongoOutputFormat", true, child)); } hj.setMapperClass((Class<? extends Mapper>) Class.forName(job.mapper, true, child)); if ((null != job.reducer) && !job.reducer.equalsIgnoreCase("null") && !job.reducer.equalsIgnoreCase("none")) { hj.setReducerClass((Class<? extends Reducer>) Class.forName(job.reducer, true, child)); } else { hj.setNumReduceTasks(0); } if ((null != job.combiner) && !job.combiner.equalsIgnoreCase("null") && !job.combiner.equalsIgnoreCase("none")) { hj.setCombinerClass((Class<? extends Reducer>) Class.forName(job.combiner, true, child)); } hj.setOutputKeyClass(Class.forName(job.outputKey, true, child)); hj.setOutputValueClass(Class.forName(job.outputValue, true, child)); hj.setJobName(job.jobtitle); if (bLocalMode) { hj.waitForCompletion(false); return "local_done"; } else { hj.submit(); String jobId = hj.getJobID().toString(); return jobId; } } catch (Exception e) { e.printStackTrace(); Thread.currentThread().setContextClassLoader(savedClassLoader); return "Error: " + HarvestExceptionUtils.createExceptionMessage(e); } finally { Thread.currentThread().setContextClassLoader(savedClassLoader); } }
From source file:com.ikanow.infinit.e.processing.custom.launcher.CustomHadoopTaskLauncher.java
License:Open Source License
@SuppressWarnings({ "unchecked", "rawtypes" }) public String runHadoopJob(CustomMapReduceJobPojo job, String tempJarLocation) throws IOException, SAXException, ParserConfigurationException { StringWriter xml = new StringWriter(); String outputCollection = job.outputCollectionTemp;// (non-append mode) if ((null != job.appendResults) && job.appendResults) outputCollection = job.outputCollection; // (append mode, write directly in....) else if (null != job.incrementalMode) job.incrementalMode = false; // (not allowed to be in incremental mode and not update mode) createConfigXML(xml, job.jobtitle, job.inputCollection, InfiniteHadoopUtils.getQueryOrProcessing(job.query, InfiniteHadoopUtils.QuerySpec.INPUTFIELDS), job.isCustomTable, job.getOutputDatabase(), job._id.toString(), outputCollection, job.mapper, job.reducer, job.combiner,/*from w w w . j a va 2s . c o m*/ InfiniteHadoopUtils.getQueryOrProcessing(job.query, InfiniteHadoopUtils.QuerySpec.QUERY), job.communityIds, job.outputKey, job.outputValue, job.arguments, job.incrementalMode, job.submitterID, job.selfMerge, job.outputCollection, job.appendResults); ClassLoader savedClassLoader = Thread.currentThread().getContextClassLoader(); URLClassLoader child = new URLClassLoader(new URL[] { new File(tempJarLocation).toURI().toURL() }, savedClassLoader); Thread.currentThread().setContextClassLoader(child); // Check version: for now, any infinit.e.data_model with an VersionTest class is acceptable boolean dataModelLoaded = true; try { URLClassLoader versionTest = new URLClassLoader(new URL[] { new File(tempJarLocation).toURI().toURL() }, null); try { Class.forName("com.ikanow.infinit.e.data_model.custom.InfiniteMongoInputFormat", true, versionTest); } catch (ClassNotFoundException e2) { //(this is fine, will use the cached version) dataModelLoaded = false; } if (dataModelLoaded) Class.forName("com.ikanow.infinit.e.data_model.custom.InfiniteMongoVersionTest", true, versionTest); } catch (ClassNotFoundException e1) { throw new RuntimeException( "This JAR is compiled with too old a version of the data-model, please recompile with Jan 2014 (rc2) onwards"); } // Now load the XML into a configuration object: Configuration config = new Configuration(); // Add the client configuration overrides: if (!bLocalMode) { String hadoopConfigPath = props_custom.getHadoopConfigPath() + "/hadoop/"; config.addResource(new Path(hadoopConfigPath + "core-site.xml")); config.addResource(new Path(hadoopConfigPath + "mapred-site.xml")); config.addResource(new Path(hadoopConfigPath + "hadoop-site.xml")); } //TESTED try { DocumentBuilderFactory dbFactory = DocumentBuilderFactory.newInstance(); DocumentBuilder dBuilder = dbFactory.newDocumentBuilder(); Document doc = dBuilder.parse(new ByteArrayInputStream(xml.toString().getBytes())); NodeList nList = doc.getElementsByTagName("property"); for (int temp = 0; temp < nList.getLength(); temp++) { Node nNode = nList.item(temp); if (nNode.getNodeType() == Node.ELEMENT_NODE) { Element eElement = (Element) nNode; String name = getTagValue("name", eElement); String value = getTagValue("value", eElement); if ((null != name) && (null != value)) { config.set(name, value); } } } } catch (Exception e) { throw new IOException(e.getMessage()); } // Some other config defaults: // (not sure if these are actually applied, or derived from the defaults - for some reason they don't appear in CDH's client config) config.set("mapred.map.tasks.speculative.execution", "false"); config.set("mapred.reduce.tasks.speculative.execution", "false"); // (default security is ignored here, have it set via HADOOP_TASKTRACKER_CONF in cloudera) // Now run the JAR file try { BasicDBObject advancedConfigurationDbo = null; try { advancedConfigurationDbo = (null != job.query) ? ((BasicDBObject) com.mongodb.util.JSON.parse(job.query)) : (new BasicDBObject()); } catch (Exception e) { advancedConfigurationDbo = new BasicDBObject(); } boolean esMode = advancedConfigurationDbo.containsField("qt") && !job.isCustomTable; if (esMode && !job.inputCollection.equals("doc_metadata.metadata")) { throw new RuntimeException( "Infinit.e Queries are only supported on doc_metadata - use MongoDB queries instead."); } config.setBoolean("mapred.used.genericoptionsparser", true); // (just stops an annoying warning from appearing) if (bLocalMode) { // local job tracker and FS mode config.set("mapred.job.tracker", "local"); config.set("fs.default.name", "local"); } else { if (bTestMode) { // run job tracker locally but FS mode remotely config.set("mapred.job.tracker", "local"); } else { // normal job tracker String trackerUrl = HadoopUtils.getXMLProperty( props_custom.getHadoopConfigPath() + "/hadoop/mapred-site.xml", "mapred.job.tracker"); config.set("mapred.job.tracker", trackerUrl); } String fsUrl = HadoopUtils.getXMLProperty( props_custom.getHadoopConfigPath() + "/hadoop/core-site.xml", "fs.default.name"); config.set("fs.default.name", fsUrl); } if (!dataModelLoaded && !(bTestMode || bLocalMode)) { // If running distributed and no data model loaded then add ourselves Path jarToCache = InfiniteHadoopUtils.cacheLocalFile("/opt/infinite-home/lib/", "infinit.e.data_model.jar", config); DistributedCache.addFileToClassPath(jarToCache, config); jarToCache = InfiniteHadoopUtils.cacheLocalFile("/opt/infinite-home/lib/", "infinit.e.processing.custom.library.jar", config); DistributedCache.addFileToClassPath(jarToCache, config); } //TESTED // Debug scripts (only if they exist), and only in non local/test mode if (!bLocalMode && !bTestMode) { try { Path scriptToCache = InfiniteHadoopUtils.cacheLocalFile("/opt/infinite-home/scripts/", "custom_map_error_handler.sh", config); config.set("mapred.map.task.debug.script", "custom_map_error_handler.sh " + job.jobtitle); config.set("mapreduce.map.debug.script", "custom_map_error_handler.sh " + job.jobtitle); DistributedCache.createSymlink(config); DistributedCache.addCacheFile(scriptToCache.toUri(), config); } catch (Exception e) { } // just carry on try { Path scriptToCache = InfiniteHadoopUtils.cacheLocalFile("/opt/infinite-home/scripts/", "custom_reduce_error_handler.sh", config); config.set("mapred.reduce.task.debug.script", "custom_reduce_error_handler.sh " + job.jobtitle); config.set("mapreduce.reduce.debug.script", "custom_reduce_error_handler.sh " + job.jobtitle); DistributedCache.createSymlink(config); DistributedCache.addCacheFile(scriptToCache.toUri(), config); } catch (Exception e) { } // just carry on } //TODO (???): TOTEST // (need to do these 2 things here before the job is created, at which point the config class has been copied across) //1) Class<?> mapperClazz = Class.forName(job.mapper, true, child); if (ICustomInfiniteInternalEngine.class.isAssignableFrom(mapperClazz)) { // Special case: internal custom engine, so gets an additional integration hook ICustomInfiniteInternalEngine preActivities = (ICustomInfiniteInternalEngine) mapperClazz .newInstance(); preActivities.preTaskActivities(job._id, job.communityIds, config, !(bTestMode || bLocalMode)); } //TESTED //2) if (job.inputCollection.equalsIgnoreCase("file.binary_shares")) { // Need to download the GridFSZip file try { Path jarToCache = InfiniteHadoopUtils.cacheLocalFile("/opt/infinite-home/lib/unbundled/", "GridFSZipFile.jar", config); DistributedCache.addFileToClassPath(jarToCache, config); } catch (Throwable t) { } // (this is fine, will already be on the classpath .. otherwise lots of other stuff will be failing all over the place!) } if (job.inputCollection.equals("records")) { InfiniteElasticsearchHadoopUtils.handleElasticsearchInput(job, config, advancedConfigurationDbo); //(won't run under 0.19 so running with "records" should cause all sorts of exceptions) } //TESTED (by hand) if (bTestMode || bLocalMode) { // If running locally, turn "snappy" off - tomcat isn't pointing its native library path in the right place config.set("mapred.map.output.compression.codec", "org.apache.hadoop.io.compress.DefaultCodec"); } // Manually specified caches List<URL> localJarCaches = InfiniteHadoopUtils.handleCacheList(advancedConfigurationDbo.get("$caches"), job, config, props_custom); Job hj = new Job(config); // (NOTE: from here, changes to config are ignored) try { if (null != localJarCaches) { if (bLocalMode || bTestMode) { Method method = URLClassLoader.class.getDeclaredMethod("addURL", new Class[] { URL.class }); method.setAccessible(true); method.invoke(child, localJarCaches.toArray()); } //TOTEST (tested logically) } Class<?> classToLoad = Class.forName(job.mapper, true, child); hj.setJarByClass(classToLoad); if (job.inputCollection.equalsIgnoreCase("filesystem")) { String inputPath = null; try { inputPath = MongoDbUtil.getProperty(advancedConfigurationDbo, "file.url"); if (!inputPath.endsWith("/")) { inputPath = inputPath + "/"; } } catch (Exception e) { } if (null == inputPath) { throw new RuntimeException("Must specify 'file.url' if reading from filesystem."); } inputPath = InfiniteHadoopUtils.authenticateInputDirectory(job, inputPath); InfiniteFileInputFormat.addInputPath(hj, new Path(inputPath + "*/*")); // (that extra bit makes it recursive) InfiniteFileInputFormat.setMaxInputSplitSize(hj, 33554432); // (32MB) InfiniteFileInputFormat.setInfiniteInputPathFilter(hj, config); hj.setInputFormatClass((Class<? extends InputFormat>) Class.forName( "com.ikanow.infinit.e.data_model.custom.InfiniteFileInputFormat", true, child)); } else if (job.inputCollection.equalsIgnoreCase("file.binary_shares")) { String[] oidStrs = null; try { String inputPath = MongoDbUtil.getProperty(advancedConfigurationDbo, "file.url"); Pattern oidExtractor = Pattern.compile("inf://share/([^/]+)"); Matcher m = oidExtractor.matcher(inputPath); if (m.find()) { oidStrs = m.group(1).split("\\s*,\\s*"); } else { throw new RuntimeException( "file.url must be in format inf://share/<oid-list>/<string>: " + inputPath); } InfiniteHadoopUtils.authenticateShareList(job, oidStrs); } catch (Exception e) { throw new RuntimeException( "Authentication error: " + e.getMessage() + ": " + advancedConfigurationDbo, e); } hj.getConfiguration().setStrings("mapred.input.dir", oidStrs); hj.setInputFormatClass((Class<? extends InputFormat>) Class.forName( "com.ikanow.infinit.e.data_model.custom.InfiniteShareInputFormat", true, child)); } else if (job.inputCollection.equals("records")) { hj.setInputFormatClass((Class<? extends InputFormat>) Class .forName("com.ikanow.infinit.e.data_model.custom.InfiniteEsInputFormat", true, child)); } else { if (esMode) { hj.setInputFormatClass((Class<? extends InputFormat>) Class.forName( "com.ikanow.infinit.e.processing.custom.utils.InfiniteElasticsearchMongoInputFormat", true, child)); } else { hj.setInputFormatClass((Class<? extends InputFormat>) Class.forName( "com.ikanow.infinit.e.data_model.custom.InfiniteMongoInputFormat", true, child)); } } if ((null != job.exportToHdfs) && job.exportToHdfs) { //TODO (INF-2469): Also, if the output key is BSON then also run as text (but output as JSON?) Path outPath = InfiniteHadoopUtils.ensureOutputDirectory(job, props_custom); if ((null != job.outputKey) && (null != job.outputValue) && job.outputKey.equalsIgnoreCase("org.apache.hadoop.io.text") && job.outputValue.equalsIgnoreCase("org.apache.hadoop.io.text")) { // (slight hack before I sort out the horrendous job class - if key/val both text and exporting to HDFS then output as Text) hj.setOutputFormatClass((Class<? extends OutputFormat>) Class .forName("org.apache.hadoop.mapreduce.lib.output.TextOutputFormat", true, child)); TextOutputFormat.setOutputPath(hj, outPath); } //TESTED else { hj.setOutputFormatClass((Class<? extends OutputFormat>) Class.forName( "org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat", true, child)); SequenceFileOutputFormat.setOutputPath(hj, outPath); } //TESTED } else { // normal case, stays in MongoDB hj.setOutputFormatClass((Class<? extends OutputFormat>) Class.forName( "com.ikanow.infinit.e.data_model.custom.InfiniteMongoOutputFormat", true, child)); } hj.setMapperClass((Class<? extends Mapper>) mapperClazz); String mapperOutputKeyOverride = advancedConfigurationDbo.getString("$mapper_key_class", null); if (null != mapperOutputKeyOverride) { hj.setMapOutputKeyClass(Class.forName(mapperOutputKeyOverride)); } //TESTED String mapperOutputValueOverride = advancedConfigurationDbo.getString("$mapper_value_class", null); if (null != mapperOutputValueOverride) { hj.setMapOutputValueClass(Class.forName(mapperOutputValueOverride)); } //TESTED if ((null != job.reducer) && !job.reducer.startsWith("#") && !job.reducer.equalsIgnoreCase("null") && !job.reducer.equalsIgnoreCase("none")) { hj.setReducerClass((Class<? extends Reducer>) Class.forName(job.reducer, true, child)); // Variable reducers: if (null != job.query) { try { hj.setNumReduceTasks(advancedConfigurationDbo.getInt("$reducers", 1)); } catch (Exception e) { try { // (just check it's not a string that is a valid int) hj.setNumReduceTasks( Integer.parseInt(advancedConfigurationDbo.getString("$reducers", "1"))); } catch (Exception e2) { } } } //TESTED } else { hj.setNumReduceTasks(0); } if ((null != job.combiner) && !job.combiner.startsWith("#") && !job.combiner.equalsIgnoreCase("null") && !job.combiner.equalsIgnoreCase("none")) { hj.setCombinerClass((Class<? extends Reducer>) Class.forName(job.combiner, true, child)); } hj.setOutputKeyClass(Class.forName(job.outputKey, true, child)); hj.setOutputValueClass(Class.forName(job.outputValue, true, child)); hj.setJobName(job.jobtitle); currJobName = job.jobtitle; } catch (Error e) { // (messing about with class loaders = lots of chances for errors!) throw new RuntimeException(e.getMessage(), e); } if (bTestMode || bLocalMode) { hj.submit(); currThreadId = null; Logger.getRootLogger().addAppender(this); currLocalJobId = hj.getJobID().toString(); currLocalJobErrs.setLength(0); while (!hj.isComplete()) { Thread.sleep(1000); } Logger.getRootLogger().removeAppender(this); if (hj.isSuccessful()) { if (this.currLocalJobErrs.length() > 0) { return "local_done: " + this.currLocalJobErrs.toString(); } else { return "local_done"; } } else { return "Error: " + this.currLocalJobErrs.toString(); } } else { hj.submit(); String jobId = hj.getJobID().toString(); return jobId; } } catch (Exception e) { e.printStackTrace(); Thread.currentThread().setContextClassLoader(savedClassLoader); return "Error: " + InfiniteHadoopUtils.createExceptionMessage(e); } finally { Thread.currentThread().setContextClassLoader(savedClassLoader); } }
From source file:com.impetus.code.examples.hadoop.cassandra.wordcount.WordCount.java
License:Apache License
public int run(String[] args) throws Exception { String outputReducerType = "cassandra"; if (args != null && args[0].startsWith(OUTPUT_REDUCER_VAR)) { String[] s = args[0].split("="); if (s != null && s.length == 2) outputReducerType = s[1];//from w ww. ja va2 s .co m } logger.info("output reducer type: " + outputReducerType); for (int i = 0; i < WordCountSetup.TEST_COUNT; i++) { String columnName = "text" + i; getConf().set(CONF_COLUMN_NAME, columnName); Job job = new Job(getConf(), "wordcount"); job.setJarByClass(WordCount.class); job.setMapperClass(TokenizerMapper.class); if (outputReducerType.equalsIgnoreCase("filesystem")) { job.setCombinerClass(ReducerToFilesystem.class); job.setReducerClass(ReducerToFilesystem.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); FileOutputFormat.setOutputPath(job, new Path(OUTPUT_PATH_PREFIX + i)); } else { job.setReducerClass(ReducerToCassandra.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(IntWritable.class); job.setOutputKeyClass(ByteBuffer.class); job.setOutputValueClass(List.class); job.setOutputFormatClass(ColumnFamilyOutputFormat.class); ConfigHelper.setOutputColumnFamily(job.getConfiguration(), KEYSPACE, OUTPUT_COLUMN_FAMILY); } job.setInputFormatClass(ColumnFamilyInputFormat.class); ConfigHelper.setRpcPort(job.getConfiguration(), "9160"); ConfigHelper.setInitialAddress(job.getConfiguration(), "localhost"); ConfigHelper.setPartitioner(job.getConfiguration(), "org.apache.cassandra.dht.RandomPartitioner"); ConfigHelper.setInputColumnFamily(job.getConfiguration(), KEYSPACE, INPUT_COLUMN_FAMILY); SlicePredicate predicate = new SlicePredicate() .setColumn_names(Arrays.asList(ByteBufferUtil.bytes(columnName))); ConfigHelper.setInputSlicePredicate(job.getConfiguration(), predicate); job.waitForCompletion(true); } return 0; }
From source file:com.impetus.code.examples.hadoop.cassandra.wordcount.WordCountCounters.java
License:Apache License
public int run(String[] args) throws Exception { Job job = new Job(getConf(), "wordcountcounters"); job.setJarByClass(WordCountCounters.class); job.setMapperClass(SumMapper.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(LongWritable.class); FileOutputFormat.setOutputPath(job, new Path(OUTPUT_PATH_PREFIX)); job.setInputFormatClass(ColumnFamilyInputFormat.class); ConfigHelper.setRpcPort(job.getConfiguration(), "9160"); ConfigHelper.setInitialAddress(job.getConfiguration(), "localhost"); ConfigHelper.setPartitioner(job.getConfiguration(), "org.apache.cassandra.dht.RandomPartitioner"); ConfigHelper.setInputColumnFamily(job.getConfiguration(), WordCount.KEYSPACE, WordCountCounters.COUNTER_COLUMN_FAMILY); SlicePredicate predicate = new SlicePredicate() .setSlice_range(new SliceRange().setStart(ByteBufferUtil.EMPTY_BYTE_BUFFER) .setFinish(ByteBufferUtil.EMPTY_BYTE_BUFFER).setCount(100)); ConfigHelper.setInputSlicePredicate(job.getConfiguration(), predicate); job.waitForCompletion(true);/*from ww w . j ava 2 s.c o m*/ return 0; }
From source file:com.impetus.code.examples.hadoop.mapred.earthquake.EarthQuakeAnalyzer.java
License:Apache License
public static void main(String[] args) throws Throwable { Job job = new Job(); job.setJarByClass(EarthQuakeAnalyzer.class); FileInputFormat.addInputPath(job, new Path("src/main/resources/eq/input")); FileOutputFormat.setOutputPath(job, new Path("src/main/resources/eq/output")); job.setMapperClass(EarthQuakeMapper.class); job.setReducerClass(EarthQuakeReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); System.exit(job.waitForCompletion(true) ? 0 : 1); }
From source file:com.impetus.code.examples.hadoop.mapred.weather.MaxTemp.java
License:Apache License
public static void main(String[] args) throws Exception { if (args.length != 2) { System.err.println("Usage: MaxTemperature <input path> <output path>"); System.exit(-1);/*from ww w . j a v a 2 s . co m*/ } Job job = new Job(); job.setJarByClass(MaxTemp.class); job.setJobName("Max temperature"); FileInputFormat.addInputPath(job, new Path(args[0])); FileOutputFormat.setOutputPath(job, new Path(args[1])); job.setMapperClass(MaxTempMapper.class); job.setReducerClass(MaxTempReducer.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(IntWritable.class); }
From source file:com.inmobi.conduit.distcp.tools.DistCp.java
License:Apache License
/** * Create Job object for submitting it, with all the configuration * * @return Reference to job object.// w w w . ja v a2 s . co m * @throws IOException - Exception if any */ protected Job createJob() throws IOException { String jobName = "distcp"; String userChosenName = getConf().get("mapred.job.name"); if (userChosenName != null) jobName += ": " + userChosenName; Job job = new Job(getConf(), jobName); job.setInputFormatClass(DistCpUtils.getStrategy(getConf(), inputOptions)); job.setJarByClass(CopyMapper.class); configureOutputFormat(job); job.setMapperClass(CopyMapper.class); job.setReducerClass(Reducer.class); job.setMapOutputKeyClass(NullWritable.class); job.setMapOutputValueClass(Text.class); job.setOutputKeyClass(NullWritable.class); job.setOutputValueClass(Text.class); job.setOutputFormatClass(CopyOutputFormat.class); job.getConfiguration().set("mapred.map.tasks.speculative.execution", "false"); job.getConfiguration().set(DistCpConstants.CONF_LABEL_NUM_MAPS, String.valueOf(inputOptions.getMaxMaps())); if (inputOptions.getSslConfigurationFile() != null) { setupSSLConfig(job.getConfiguration()); } inputOptions.appendToConf(job.getConfiguration()); return job; }
From source file:com.inmobi.conduit.local.LocalStreamService.java
License:Apache License
protected Job createJob(Path inputPath, long totalSize) throws IOException { String jobName = getName();// w ww. java2s . c o m Configuration conf = currentCluster.getHadoopConf(); conf.set(ConduitConstants.AUDIT_ENABLED_KEY, System.getProperty(ConduitConstants.AUDIT_ENABLED_KEY)); Job job = new Job(conf); job.setJobName(jobName); // DistributedCache.addFileToClassPath(inputFormatJarDestPath, // job.getConfiguration()); job.getConfiguration().set("tmpjars", inputFormatJarDestPath.toString() + "," + auditUtilJarDestPath.toString()); LOG.debug("Adding file [" + inputFormatJarDestPath + "] to distributed cache"); job.setInputFormatClass(UniformSizeInputFormat.class); Class<? extends Mapper<Text, FileStatus, NullWritable, Text>> mapperClass = getMapperClass(); job.setJarByClass(mapperClass); job.setMapperClass(mapperClass); job.setMapOutputKeyClass(NullWritable.class); job.setMapOutputValueClass(Text.class); job.setOutputKeyClass(NullWritable.class); job.setOutputValueClass(Text.class); // setting identity reducer job.setReducerClass(Reducer.class); job.setOutputFormatClass(TextOutputFormat.class); TextOutputFormat.setOutputPath(job, tmpCounterOutputPath); job.getConfiguration().set("mapred.map.tasks.speculative.execution", "false"); job.getConfiguration().set(LOCALSTREAM_TMP_PATH, tmpPath.toString()); job.getConfiguration().set(SRC_FS_DEFAULT_NAME_KEY, srcCluster.getHadoopConf().get(FS_DEFAULT_NAME_KEY)); // set configurations needed for UniformSizeInputFormat int numMaps = getNumMapsForJob(totalSize); job.getConfiguration().setInt(DistCpConstants.CONF_LABEL_NUM_MAPS, numMaps); job.getConfiguration().setLong(DistCpConstants.CONF_LABEL_TOTAL_BYTES_TO_BE_COPIED, totalSize); job.getConfiguration().set(DistCpConstants.CONF_LABEL_LISTING_FILE_PATH, inputPath.toString()); LOG.info("Expected number of maps [" + numMaps + "] Total data size [" + totalSize + "]"); return job; }