List of usage examples for org.apache.hadoop.conf Configuration Configuration
public Configuration()
From source file:at.illecker.hama.hybrid.examples.onlinecf.OnlineCF.java
License:Apache License
@Override public boolean load(String path, boolean lazy) { this.m_isLazyLoadModel = lazy; this.m_modelPath = path; if (lazy == false) { Path dataPath = new Path(m_modelPath); Configuration conf = new Configuration(); try {//from w w w.j av a 2s .co m FileSystem fs = dataPath.getFileSystem(conf); LinkedList<Path> files = new LinkedList<Path>(); if (!fs.exists(dataPath)) { this.m_isLazyLoadModel = false; this.m_modelPath = null; return false; } if (!fs.isFile(dataPath)) { for (int i = 0; i < 100000; i++) { Path partFile = new Path( m_modelPath + "/part-" + String.valueOf(100000 + i).substring(1, 6)); if (fs.exists(partFile)) { files.add(partFile); } else { break; } } } else { files.add(dataPath); } LOG.info("loading model from " + path); for (Path file : files) { SequenceFile.Reader reader = new SequenceFile.Reader(fs, file, conf); Text key = new Text(); PipesVectorWritable value = new PipesVectorWritable(); String strKey = null; Long actualKey = null; String firstSymbol = null; while (reader.next(key, value) != false) { strKey = key.toString(); firstSymbol = strKey.substring(0, 1); try { actualKey = Long.valueOf(strKey.substring(1)); } catch (Exception e) { actualKey = new Long(0); } if (firstSymbol.equals(OnlineCF.DFLT_MODEL_ITEM_DELIM)) { // LOG.info("loaded itemId: " + actualKey + " itemVector: " // + value.getVector()); m_modelItemFactorizedValues.put(actualKey, new PipesVectorWritable(value)); } else if (firstSymbol.equals(OnlineCF.DFLT_MODEL_USER_DELIM)) { // LOG.info("loaded userId: " + actualKey + " userVector: " // + value.getVector()); m_modelUserFactorizedValues.put(actualKey, new PipesVectorWritable(value)); } else { // unknown continue; } } reader.close(); } LOG.info("loaded: " + m_modelUserFactorizedValues.size() + " users, " + m_modelItemFactorizedValues.size() + " items"); // for (Long user : m_modelUserFactorizedValues.keySet()) { // LOG.info("userId: " + user + " userVector: " // + m_modelUserFactorizedValues.get(user)); // } // for (Long item : m_modelItemFactorizedValues.keySet()) { // LOG.info("itemId: " + item + " itemVector: " // + m_modelItemFactorizedValues.get(item)); // } } catch (Exception e) { e.printStackTrace(); this.m_isLazyLoadModel = false; this.m_modelPath = null; return false; } } return true; }
From source file:at.illecker.hama.hybrid.examples.onlinecf.OnlineCFHybridBenchmark.java
License:Apache License
@Override protected void setUp() throws Exception { m_conf = new Configuration(); // Try to load Hadoop configuration String HADOOP_HOME = System.getenv("HADOOP_HOME"); String HADOOP_INSTALL = System.getenv("HADOOP_INSTALL"); if ((HADOOP_HOME != null) || (HADOOP_INSTALL != null) && (!m_runLocally)) { String HADOOP = ((HADOOP_HOME != null) ? HADOOP_HOME : HADOOP_INSTALL); m_conf.addResource(new Path(HADOOP, "src/core/core-default.xml")); m_conf.addResource(new Path(HADOOP, "src/hdfs/hdfs-default.xml")); m_conf.addResource(new Path(HADOOP, "src/mapred/mapred-default.xml")); m_conf.addResource(new Path(HADOOP, "conf/core-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/hdfs-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/mapred-site.xml")); // System.out.println("Loaded Hadoop configuration from " + HADOOP); try {//from w w w .j a va2s . co m // Connect to HDFS Filesystem FileSystem.get(m_conf); } catch (Exception e) { // HDFS not reachable run Benchmark locally m_conf = new Configuration(); m_runLocally = true; } // System.out.println("Run Benchmark local: " + m_runLocally); } // Try to load Hama configuration String HAMA_HOME = System.getenv("HAMA_HOME"); String HAMA_INSTALL = System.getenv("HAMA_INSTALL"); if ((HAMA_HOME != null) || (HAMA_INSTALL != null) && (!m_runLocally)) { String HAMA = ((HAMA_HOME != null) ? HAMA_HOME : HAMA_INSTALL); m_conf.addResource(new Path(HAMA, "conf/hama-default.xml")); m_conf.addResource(new Path(HAMA, "conf/hama-site.xml")); // System.out.println("Loaded Hama configuration from " + HAMA); } // Setup OnlineCF config variables m_conf.setBoolean(OnlineCFTrainHybridBSP.CONF_DEBUG, false); m_conf.setBoolean("hama.pipes.logging", false); // Set GPU blockSize and gridSize m_conf.set(OnlineCFTrainHybridBSP.CONF_BLOCKSIZE, "" + BLOCK_SIZE); m_conf.set(OnlineCFTrainHybridBSP.CONF_GRIDSIZE, "" + GRID_SIZE); int numGpuBspTask = 0; // CPU vs GPU iterations benchmark // Plot 1 and 2 // if (type == CalcType.GPU) { // bspTaskNum = 1; // numGpuBspTask = 1; // GPUWorkload = 100; // } // CPU + GPU Hybrid benchmark // Plot 3 if (bspTaskNum == maxTaskNum) { numGpuBspTask = 1; // GPUWorkload = 75; } else { numGpuBspTask = 0; } // Set CPU tasks m_conf.setInt("bsp.peers.num", bspTaskNum); // Set GPU tasks m_conf.setInt("bsp.peers.gpu.num", numGpuBspTask); m_conf.setInt(OnlineCF.CONF_ITERATION_COUNT, iteration); m_conf.setInt(OnlineCF.CONF_MATRIX_RANK, matrixRank); m_conf.setInt(OnlineCF.CONF_SKIP_COUNT, skipCount); Path preferencesIn = new Path(CONF_INPUT_DIR, "preferences_in.seq"); if (!m_useInputFile) { // Generate random input data m_testPrefs = generateRandomInputData(m_conf, FileSystem.get(m_conf), CONF_INPUT_DIR, bspTaskNum, numGpuBspTask, n, m, percentNonZeroValues, GPUWorkload, m_maxTestPrefs); } else { // Convert MovieLens input data // parse inputFile and return first entries for testing m_testPrefs = convertInputData(m_conf, FileSystem.get(m_conf), CONF_INPUT_DIR, preferencesIn, m_movieLensInputFile, "::", m_maxTestPrefs); } // Debug output // System.out.println("CalcType: " + type); System.out.println("CONF_TMP_DIR: " + CONF_TMP_DIR.toString()); System.out.println("NumBspTask: " + m_conf.getInt("bsp.peers.num", 0) + " NumGpuBspTask: " + m_conf.getInt("bsp.peers.gpu.num", 0)); if (!m_useInputFile) { System.out.println("n: " + n + " m: " + m + " percentNonZeroValues: " + percentNonZeroValues); } else { System.out.println("Use inputFile: " + m_movieLensInputFile); } System.out.println("matrixRank: " + matrixRank + " iterations: " + iteration); }
From source file:at.illecker.hama.hybrid.examples.piestimator.PiEstimatorHybridBenchmark.java
License:Apache License
@Override protected void setUp() throws Exception { m_conf = new Configuration(); // Try to load Hadoop configuration String HADOOP_HOME = System.getenv("HADOOP_HOME"); String HADOOP_INSTALL = System.getenv("HADOOP_INSTALL"); if ((HADOOP_HOME != null) || (HADOOP_INSTALL != null) && (!m_runLocally)) { String HADOOP = ((HADOOP_HOME != null) ? HADOOP_HOME : HADOOP_INSTALL); m_conf.addResource(new Path(HADOOP, "src/core/core-default.xml")); m_conf.addResource(new Path(HADOOP, "src/hdfs/hdfs-default.xml")); m_conf.addResource(new Path(HADOOP, "src/mapred/mapred-default.xml")); m_conf.addResource(new Path(HADOOP, "conf/core-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/hdfs-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/mapred-site.xml")); // System.out.println("Loaded Hadoop configuration from " + HADOOP); try {/*from w w w.j ava2s. co m*/ // Connect to HDFS Filesystem FileSystem.get(m_conf); } catch (Exception e) { // HDFS not reachable run Benchmark locally m_conf = new Configuration(); m_runLocally = true; } // System.out.println("Run Benchmark local: " + m_runLocally); } // Try to load Hama configuration String HAMA_HOME = System.getenv("HAMA_HOME"); String HAMA_INSTALL = System.getenv("HAMA_INSTALL"); if ((HAMA_HOME != null) || (HAMA_INSTALL != null) && (!m_runLocally)) { String HAMA = ((HAMA_HOME != null) ? HAMA_HOME : HAMA_INSTALL); m_conf.addResource(new Path(HAMA, "conf/hama-default.xml")); m_conf.addResource(new Path(HAMA, "conf/hama-site.xml")); // System.out.println("Loaded Hama configuration from " + HAMA); } // calculate total sampling size m_totalIterations = (long) 1024 * (long) 14 * (long) 1000 * n; int numGpuBspTask = 0; // Used only for Plot 1 - CPU vs GPU comparison // if (type == CalcType.GPU) { // bspTaskNum = 1; // numGpuBspTask = 1; // GPUWorkload = 100; // } // Used only for Plot 2 - CPU + GPU Hybrid benchmark if (bspTaskNum == maxBspTaskNum) { numGpuBspTask = 1; GPUWorkload = 95; } else { numGpuBspTask = 0; } // Set CPU tasks m_conf.setInt("bsp.peers.num", bspTaskNum); // Set GPU tasks m_conf.setInt("bsp.peers.gpu.num", numGpuBspTask); m_conf.setInt(PiEstimatorHybridBSP.CONF_BLOCKSIZE, m_blockSize); m_conf.setInt(PiEstimatorHybridBSP.CONF_GRIDSIZE, m_gridSize); m_conf.setLong(PiEstimatorHybridBSP.CONF_ITERATIONS, m_totalIterations); m_conf.setInt(PiEstimatorHybridBSP.CONF_GPU_PERCENTAGE, GPUWorkload); m_conf.setBoolean(PiEstimatorHybridBSP.CONF_DEBUG, false); m_conf.setBoolean(PiEstimatorHybridBSP.CONF_TIME, false); // Debug output System.out.println("Benchmark PiEstimatorHybridBSP[blockSize=" + m_blockSize + ",gridSize=" + m_gridSize + "] n=" + n + ",bspTaskNum=" + bspTaskNum + ",GpuBspTaskNum=" + numGpuBspTask + ",GPUWorkload=" + GPUWorkload + ",totalSamples=" + m_totalIterations); System.out.println("CONF_TMP_DIR: " + CONF_TMP_DIR.toString()); }
From source file:at.illecker.hama.rootbeer.examples.matrixmultiplication.MatrixMultiplicationBenchmark.java
License:Apache License
@Override protected void setUp() throws Exception { m_conf = new Configuration(); // Try to load Hadoop configuration String HADOOP_HOME = System.getenv("HADOOP_HOME"); String HADOOP_INSTALL = System.getenv("HADOOP_INSTALL"); if ((HADOOP_HOME != null) || (HADOOP_INSTALL != null) && (!m_runLocally)) { String HADOOP = ((HADOOP_HOME != null) ? HADOOP_HOME : HADOOP_INSTALL); m_conf.addResource(new Path(HADOOP, "src/core/core-default.xml")); m_conf.addResource(new Path(HADOOP, "src/hdfs/hdfs-default.xml")); m_conf.addResource(new Path(HADOOP, "src/mapred/mapred-default.xml")); m_conf.addResource(new Path(HADOOP, "conf/core-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/hdfs-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/mapred-site.xml")); // System.out.println("Loaded Hadoop configuration from " + HADOOP); try {/* w w w . java 2 s. c om*/ // Connect to HDFS Filesystem FileSystem.get(m_conf); } catch (Exception e) { // HDFS not reachable run Benchmark locally m_conf = new Configuration(); m_runLocally = true; } // System.out.println("Run Benchmark local: " + m_runLocally); } // Try to load Hama configuration String HAMA_HOME = System.getenv("HAMA_HOME"); String HAMA_INSTALL = System.getenv("HAMA_INSTALL"); if ((HAMA_HOME != null) || (HAMA_INSTALL != null) && (!m_runLocally)) { String HAMA = ((HAMA_HOME != null) ? HAMA_HOME : HAMA_INSTALL); m_conf.addResource(new Path(HAMA, "conf/hama-default.xml")); m_conf.addResource(new Path(HAMA, "conf/hama-site.xml")); // System.out.println("Loaded Hama configuration from " + HAMA); } // Setup outputs m_OUTPUT_DIR_PATH = new Path(OUTPUT_DIR + "/bench_" + System.currentTimeMillis()); System.out.println("OUTPUT_DIR_PATH: " + m_OUTPUT_DIR_PATH); m_MATRIX_A_PATH = new Path(m_OUTPUT_DIR_PATH + "/MatrixA.seq"); m_MATRIX_B_PATH = new Path(m_OUTPUT_DIR_PATH + "/MatrixB.seq"); m_MATRIX_C_PATH = new Path(m_OUTPUT_DIR_PATH + "/MatrixC.seq"); m_MATRIX_D_PATH = new Path(m_OUTPUT_DIR_PATH + "/MatrixD.seq"); m_blockSize = MatrixMultiplicationBSPGpu.BLOCK_SIZE; m_gridSize = MatrixMultiplicationBSPGpu.GRID_SIZE; System.out.println("Benchmark MatrixMultiplication " + type + " [blockSize=" + m_blockSize + ",gridSize=" + m_gridSize + "] " + n + " x " + n + " matrix"); // Create random DistributedRowMatrix DistributedRowMatrix.createRandomDistributedRowMatrix(m_conf, n, n, new Random(42L), m_MATRIX_A_PATH, false); DistributedRowMatrix.createRandomDistributedRowMatrix(m_conf, n, n, new Random(1337L), m_MATRIX_B_PATH, (type == CalcType.CPU) ? true : false); // Load DistributedRowMatrix a and b m_matrixA = new DistributedRowMatrix(m_MATRIX_A_PATH, m_OUTPUT_DIR_PATH, n, n); m_matrixB = new DistributedRowMatrix(m_MATRIX_B_PATH, m_OUTPUT_DIR_PATH, n, n); m_matrixA.setConf(m_conf); m_matrixB.setConf(m_conf); }
From source file:at.illecker.hama.rootbeer.examples.piestimator.PiEstimatorBenchmark.java
License:Apache License
@Override protected void setUp() throws Exception { m_conf = new Configuration(); // Try to load Hadoop configuration String HADOOP_HOME = System.getenv("HADOOP_HOME"); String HADOOP_INSTALL = System.getenv("HADOOP_INSTALL"); if ((HADOOP_HOME != null) || (HADOOP_INSTALL != null) && (!m_runLocally)) { String HADOOP = ((HADOOP_HOME != null) ? HADOOP_HOME : HADOOP_INSTALL); m_conf.addResource(new Path(HADOOP, "src/core/core-default.xml")); m_conf.addResource(new Path(HADOOP, "src/hdfs/hdfs-default.xml")); m_conf.addResource(new Path(HADOOP, "src/mapred/mapred-default.xml")); m_conf.addResource(new Path(HADOOP, "conf/core-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/hdfs-site.xml")); m_conf.addResource(new Path(HADOOP, "conf/mapred-site.xml")); // System.out.println("Loaded Hadoop configuration from " + HADOOP); try {//from w ww. j a v a2 s . c om // Connect to HDFS Filesystem FileSystem.get(m_conf); } catch (Exception e) { // HDFS not reachable run Benchmark locally m_conf = new Configuration(); m_runLocally = true; } // System.out.println("Run Benchmark local: " + m_runLocally); } // Try to load Hama configuration String HAMA_HOME = System.getenv("HAMA_HOME"); String HAMA_INSTALL = System.getenv("HAMA_INSTALL"); if ((HAMA_HOME != null) || (HAMA_INSTALL != null) && (!m_runLocally)) { String HAMA = ((HAMA_HOME != null) ? HAMA_HOME : HAMA_INSTALL); m_conf.addResource(new Path(HAMA, "conf/hama-default.xml")); m_conf.addResource(new Path(HAMA, "conf/hama-site.xml")); // System.out.println("Loaded Hama configuration from " + HAMA); } // Setup outputs m_OUTPUT_DIR_PATH = new Path(OUTPUT_DIR + "/bench_" + System.currentTimeMillis()); System.out.println("OUTPUT_DIR_PATH: " + m_OUTPUT_DIR_PATH); m_blockSize = PiEstimatorGpuBSP.blockSize; m_gridSize = PiEstimatorGpuBSP.gridSize; m_totalIterations = (long) m_blockSize * (long) m_gridSize * n; System.out.println("Benchmark PiEstimator[blockSize=" + m_blockSize + ",gridSize=" + m_gridSize + "] n=" + n + ", totalSamples=" + m_totalIterations); }
From source file:at.illecker.hama.rootbeer.examples.util.RandomGraphGenerator.java
License:Apache License
public static void main(String[] args) throws Exception { if (args.length != 4) { System.out.println(//from w w w. ja v a2 s. c om "USAGE: <Number of vertices> <Number of edges per vertex> <Number of partitions> <Outpath>"); return; } System.out.println(Arrays.toString(args)); Configuration conf = new Configuration(); conf.setInt("hama.num.vertices", Integer.parseInt(args[0])); conf.setInt("hama.num.partitions", Integer.parseInt(args[2])); conf.setInt("number.edges", Integer.parseInt(args[1])); Job job = new Job(conf); Path generated = new Path(new Path(args[3]).getParent(), "generated"); FileOutputFormat.setOutputPath(job, generated); FileSystem.get(conf).delete(generated, true); job.setJobName("RangeWriter"); job.setJarByClass(SortGenMapper.class); job.setMapperClass(SortGenMapper.class); job.setNumReduceTasks(0); job.setOutputFormatClass(SequenceFileOutputFormat.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(NullWritable.class); job.setInputFormatClass(RangeInputFormat.class); job.waitForCompletion(true); conf.setInt("max.id", Integer.valueOf(args[0])); job = new Job(conf); FileOutputFormat.setOutputPath(job, new Path(args[3])); FileSystem.get(conf).delete(new Path(args[3]), true); job.setJobName("Random Vertex Writer"); FileInputFormat.addInputPath(job, generated); job.setJarByClass(RandomMapper.class); job.setMapperClass(RandomMapper.class); job.setReducerClass(Reducer.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(Text.class); job.setNumReduceTasks(conf.getInt("hama.num.partitions", 2)); job.setPartitionerClass(HashPartitioner.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Text.class); job.setInputFormatClass(SequenceFileInputFormat.class); job.setOutputFormatClass(TextOutputFormat.class); job.waitForCompletion(true); }
From source file:audr.text.utils.FileUtils.java
License:Open Source License
/** * hadoop// ww w .j a va 2s.com * * @param localFile * * @param hadoopFile * hadoop * @return */ public static void uploadFile2HDFS(String localFile, String hadoopFile) { try { Configuration conf = new Configuration(); FileSystem src = FileSystem.getLocal(conf); FileSystem dst = FileSystem.get(conf); Path srcpath = new Path(localFile); Path dstpath = new Path(hadoopFile); FileUtil.copy(src, srcpath, dst, dstpath, false, conf); } catch (Exception e) { e.printStackTrace(); } }
From source file:audr.text.utils.FileUtils.java
License:Open Source License
/** * HDFSpath//from w w w . j a v a 2 s . c o m * * @param path */ public static void deleteFileFromHDFS(String path) { try { Configuration conf = new Configuration(); FileSystem dst = FileSystem.get(conf); Path dstpath = new Path(path); dst.delete(dstpath, true); } catch (Exception e) { e.printStackTrace(); } }
From source file:audr.text.utils.FileUtils.java
License:Open Source License
/** * /*from w w w.j a v a 2 s. com*/ * * @param path * @return true false */ public static boolean isDirectoryExist(Path path) { int existFlag = 0; try { Configuration conf = new Configuration(); FileSystem dst = FileSystem.get(conf); if (dst.exists(path)) existFlag = 1; } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } if (existFlag == 1) return true; return false; }
From source file:audr.text.utils.FileUtils.java
License:Open Source License
public static String readTextFromHDFS(Path filePath) throws IOException { // Configuration Configuration conf = new Configuration(); // DataInputStream FSDataInputStream in = null;/*from ww w . java 2 s . c o m*/ // byteFSDataInputStream byte[] line; // FileSystem fs = FileSystem.get(conf); // FileSystemopen in = fs.open(filePath); line = new byte[MAX_LENGTH]; in.read(line); return new String(line).trim(); }