List of usage examples for org.apache.hadoop.mapreduce Job getConfiguration
public Configuration getConfiguration()
From source file:com.ci.backports.avro.mapreduce.AvroOutputFormat.java
License:Apache License
/** Enable output compression using the deflate codec and specify its level.*/ public static void setDeflateLevel(Job job, int level) { FileOutputFormat.setCompressOutput(job, true); job.getConfiguration().setInt(org.apache.avro.mapred.AvroOutputFormat.DEFLATE_LEVEL_KEY, level); }
From source file:com.ci.backports.hadoop.hbase.ZHFileOutputFormat.java
License:Apache License
/** * Configure a MapReduce Job to perform an incremental load into the given * table. This/* w w w . j a va2 s .co m*/ * <ul> * <li>Inspects the table to configure a total order partitioner</li> * <li>Uploads the partitions file to the cluster and adds it to the DistributedCache</li> * <li>Sets the number of reduce tasks to match the current number of regions</li> * <li>Sets the output key/value class to match ZHFileOutputFormat's requirements</li> * <li>Sets the reducer up to perform the appropriate sorting (either KeyValueSortReducer or * ZPutSortReducer)</li> * </ul> * The user should be sure to set the map output value class to either KeyValue or Put before * running this function. */ public static void configureIncrementalLoad(Job job, HTable table) throws IOException { Configuration conf = job.getConfiguration(); job.setPartitionerClass(TotalOrderPartitioner.class); job.setOutputKeyClass(ImmutableBytesWritable.class); job.setOutputValueClass(KeyValue.class); job.setOutputFormatClass(ZHFileOutputFormat.class); // Based on the configured map output class, set the correct reducer to properly // sort the incoming values. // TODO it would be nice to pick one or the other of these formats. if (KeyValue.class.equals(job.getMapOutputValueClass())) { job.setReducerClass(KeyValueSortReducer.class); } else if (Put.class.equals(job.getMapOutputValueClass())) { job.setReducerClass(ZPutSortReducer.class); } else { LOG.warn("Unknown map output value type:" + job.getMapOutputValueClass()); } LOG.info("Looking up current regions for table " + table); List<ImmutableBytesWritable> startKeys = getRegionStartKeys(table); LOG.info("Configuring " + startKeys.size() + " reduce partitions " + "to match current region count"); job.setNumReduceTasks(startKeys.size()); Path partitionsPath = new Path(job.getWorkingDirectory(), "partitions_" + System.currentTimeMillis()); LOG.info("Writing partition information to " + partitionsPath); FileSystem fs = partitionsPath.getFileSystem(conf); writePartitions(conf, partitionsPath, startKeys); partitionsPath.makeQualified(fs); URI cacheUri; try { cacheUri = new URI(partitionsPath.toString() + "#" + TotalOrderPartitioner.DEFAULT_PATH); } catch (URISyntaxException e) { throw new IOException(e); } DistributedCache.addCacheFile(cacheUri, conf); DistributedCache.createSymlink(conf); LOG.info("Incremental table output configured."); }
From source file:com.citic.zxyjs.zwlscx.mapreduce.lib.input.HFileOutputFormatBase.java
License:Apache License
/** * Configure a MapReduce Job to perform an incremental load into the given * table. This/*from ww w. java 2 s.c o m*/ * <ul> * <li>Inspects the table to configure a total order partitioner</li> * <li>Uploads the partitions file to the cluster and adds it to the * DistributedCache</li> * <li>Sets the number of reduce tasks to match the current number of * regions</li> * <li>Sets the output key/value class to match HFileOutputFormat's * requirements</li> * <li>Sets the reducer up to perform the appropriate sorting (either * KeyValueSortReducer or PutSortReducer)</li> * </ul> * The user should be sure to set the map output value class to either * KeyValue or Put before running this function. */ public static void configureIncrementalLoad(Job job, HTable table, Class<? extends HFileOutputFormatBase> hfileOutputFormatBase) throws IOException { Configuration conf = job.getConfiguration(); job.setOutputKeyClass(ImmutableBytesWritable.class); job.setOutputValueClass(KeyValue.class); job.setOutputFormatClass(hfileOutputFormatBase); // Based on the configured map output class, set the correct reducer to // properly // sort the incoming values. // TODO it would be nice to pick one or the other of these formats. if (KeyValue.class.equals(job.getMapOutputValueClass())) { job.setReducerClass(KeyValueSortReducer.class); } else if (Put.class.equals(job.getMapOutputValueClass())) { job.setReducerClass(PutSortReducer.class); } else if (Text.class.equals(job.getMapOutputValueClass())) { job.setReducerClass(TextSortReducer.class); } else { LOG.warn("Unknown map output value type:" + job.getMapOutputValueClass()); } conf.setStrings("io.serializations", conf.get("io.serializations"), MutationSerialization.class.getName(), ResultSerialization.class.getName(), KeyValueSerialization.class.getName()); // Use table's region boundaries for TOP split points. LOG.info("Looking up current regions for table " + Bytes.toString(table.getTableName())); List<ImmutableBytesWritable> startKeys = getRegionStartKeys(table); LOG.info("Configuring " + startKeys.size() + " reduce partitions " + "to match current region count"); job.setNumReduceTasks(startKeys.size()); configurePartitioner(job, startKeys); // Set compression algorithms based on column families configureCompression(table, conf); configureBloomType(table, conf); configureBlockSize(table, conf); // TableMapReduceUtil.addDependencyJars(job); TableMapReduceUtil.initCredentials(job); LOG.info("Incremental table " + Bytes.toString(table.getTableName()) + " output configured."); }
From source file:com.citic.zxyjs.zwlscx.mapreduce.lib.input.HFileOutputFormatBase.java
License:Apache License
/** * Configure <code>job</code> with a TotalOrderPartitioner, partitioning * against <code>splitPoints</code>. Cleans up the partitions file after job * exists./* w w w. j av a 2 s. c o m*/ */ static void configurePartitioner(Job job, List<ImmutableBytesWritable> splitPoints) throws IOException { // create the partitions file FileSystem fs = FileSystem.get(job.getConfiguration()); Path partitionsPath = new Path("/tmp", "partitions_" + UUID.randomUUID()); fs.makeQualified(partitionsPath); fs.deleteOnExit(partitionsPath); writePartitions(job.getConfiguration(), partitionsPath, splitPoints); // configure job to use it job.setPartitionerClass(TotalOrderPartitioner.class); TotalOrderPartitioner.setPartitionFile(job.getConfiguration(), partitionsPath); }
From source file:com.cloudera.accumulo.upgrade.compatibility.DataCompatibilityLoad.java
License:Open Source License
@Override public int run(String[] args) throws Exception { final String jobName = this.getClass().getName(); options.parseArgs(jobName, args);/*from w ww.java2 s .com*/ final Job job = new Job(getConf(), jobName); if (-1 == options.test.numRows) { options.test.numRows = job.getConfiguration().getInt("mapred.map.tasks", DataCompatibilityTestCli.DEFAULT_NUM_ROWS); } job.setJarByClass(this.getClass()); job.setInputFormatClass(DataLoadInputFormat.class); DataLoadInputFormat.setTabletServers(job, options.connection.getConnector().instanceOperations().getTabletServers()); DataLoadInputFormat.setNumRows(job, options.test.numRows); DataLoadInputFormat.setNumQualifiersPerFamily(job, options.test.qualifiers); job.getConfiguration().set(VISIBILITY, new String(options.visibility.visibility.getExpression(), "UTF-8")); final TableOperations ops = options.connection.getConnector().tableOperations(); final List<String> names = options.test.getTableNamesAndConfigureThem(ops); for (String name : names) { final int numSplits = ops.getSplits(name, options.test.numRows).size(); if (options.test.numRows > numSplits) { log.info("adding splits to table '" + name + "', to bring it from " + numSplits + " to " + options.test.numRows + "."); final SortedSet<Text> splits = new TreeSet<Text>(); // for cases where we're adding way more splits than there are currently possible servers to handle them, do a pre-pre-split // N.B. If we've just created this table, there will be 0 splits because we'll just have the initial tablet. if (0 == numSplits || options.test.numRows / numSplits > 10) { log.info("splitting in two waves due to the number of splits we need to add."); // TODO turtles all the way down. final int prepre = options.test.numRows / (0 == numSplits ? 10 : numSplits * 10); for (int i = 0; i < prepre; i++) { splits.add(new Text(new StringBuilder(Long.toString(i)).reverse().toString())); } ops.addSplits(name, splits); log.debug("delay 30s for splits to get assigned off host."); try { Thread.currentThread().sleep(30 * 1000); } catch (InterruptedException exception) { log.warn("interrupted from sleep early."); } splits.clear(); } for (int i = 0; i < options.test.numRows; i++) { splits.add(new Text(new StringBuilder(Long.toString(i)).reverse().toString())); } ops.addSplits(name, splits); } } log.debug("delay 30s for splits to get assigned off host."); try { Thread.currentThread().sleep(30 * 1000); } catch (InterruptedException exception) { log.warn("interrupted from sleep early."); } job.getConfiguration().setStrings(OUTPUT_TABLE_NAMES, names.toArray(new String[0])); job.setMapperClass(DataLoadMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(Mutation.class); job.setNumReduceTasks(0); log.info("launching map-only job to insert " + options.test.numRows + " rows of " + (FAMILIES.length * options.test.qualifiers) + " cells each into each of the tables " + names); options.output.useAccumuloOutputFormat(job); job.waitForCompletion(true); return job.isSuccessful() ? 0 : 1; }
From source file:com.cloudera.accumulo.upgrade.util.MapreduceInputCli.java
License:Open Source License
/** * Iff you use offline mode, you have to call close() when you're done or manually delete the cloned table yourself. */// w w w .jav a 2s .c o m public void useAccumuloInputFormat(Job job, String table, boolean offline) throws IOException, AccumuloException, AccumuloSecurityException, TableExistsException, TableNotFoundException { job.setInputFormatClass(AccumuloInputFormat.class); /* XXX Need to use a method that exists in 1.4 adn 1.5 :( */ Configuration configuration = job.getConfiguration(); AccumuloInputFormat.setZooKeeperInstance(job, connection.instance, connection.zookeepers); final TableOperations ops = connection.getConnector().tableOperations(); String scan = table; if (offline) { Random random = new Random(); scan = table + "_" + String.format("%016x", Math.abs(random.nextLong())); ops.clone(table, scan, true, Collections.<String, String>emptyMap(), Collections.<String>emptySet()); try { ops.offline(scan); } finally { clones.add(scan); } AccumuloInputFormat.setOfflineTableScan(job, true); } PasswordToken token = new PasswordToken(connection.password); AccumuloInputFormat.setConnectorInfo(job, connection.principal, token); AccumuloInputFormat.setInputTableName(job, scan); AccumuloInputFormat.setScanAuthorizations(job, connection.auths); if (0 < maxMaps) { // set up ranges try { Set<Range> ranges = ops.splitRangeByTablets(table, new Range(), maxMaps); AccumuloInputFormat.setRanges(job, ranges); AccumuloInputFormat.setAutoAdjustRanges(job, false); } catch (Exception e) { throw new IOException(e); } } }
From source file:com.cloudera.bigdata.analysis.dataload.io.SplitableInputFormat.java
License:Apache License
public static void setReaderClass(Job job, String className) { job.getConfiguration().set(RECORD_READER, className); }
From source file:com.cloudera.bigdata.analysis.dataload.io.SplitableInputFormat.java
License:Apache License
public static void setEncoding(Job job, String encoding) { job.getConfiguration().set(ENCODING, encoding); }
From source file:com.cloudera.bigdata.analysis.dataload.io.SplitableInputFormat.java
License:Apache License
public static void setDelimiter(Job job, String delimiter) { job.getConfiguration().set(DELIMITER, delimiter); }
From source file:com.cloudera.crunch.impl.mr.collect.PGroupedTableImpl.java
License:Open Source License
public void configureShuffle(Job job) { ptype.configureShuffle(job, groupingOptions); if (groupingOptions != null && groupingOptions.getNumReducers() <= 0) { int bytesPerTask = job.getConfiguration().getInt("crunch.bytes.per.reduce.task", (1000 * 1000 * 1000)); int numReduceTasks = 1 + (int) (getSize() / bytesPerTask); job.setNumReduceTasks(numReduceTasks); LOG.info(String.format("Setting num reduce tasks to %d", numReduceTasks)); }/* www .ja v a2 s. c o m*/ }