org.apache.hadoop.hbase.HColumnDescriptor#setCompressionType

Source File: HBaseBasedAuditRepository.java From atlas with Apache License 2.0

6 votes

private void createTableIfNotExists() throws AtlasException {
    Admin admin = null;
    try {
        admin = connection.getAdmin();
        LOG.info("Checking if table {} exists", tableName.getNameAsString());
        if (!admin.tableExists(tableName)) {
            LOG.info("Creating table {}", tableName.getNameAsString());
            HTableDescriptor tableDescriptor = new HTableDescriptor(tableName);
            HColumnDescriptor columnFamily = new HColumnDescriptor(COLUMN_FAMILY);
            columnFamily.setMaxVersions(1);
            columnFamily.setDataBlockEncoding(DataBlockEncoding.FAST_DIFF);
            columnFamily.setCompressionType(Compression.Algorithm.GZ);
            columnFamily.setBloomFilterType(BloomType.ROW);
            tableDescriptor.addFamily(columnFamily);
            admin.createTable(tableDescriptor);
        } else {
            LOG.info("Table {} exists", tableName.getNameAsString());
        }
    } catch (IOException e) {
        throw new AtlasException(e);
    } finally {
        close(admin);
    }
}

Source File: CreateTable.java From examples with Apache License 2.0

6 votes

public static void main(String[] args) throws MasterNotRunningException,
    ZooKeeperConnectionException, IOException {
  try (Connection connection = ConnectionFactory.createConnection();
      Admin admin = connection.getAdmin();) {
    LOG.info("Starting table creation");
    // tag::CREATE[]
    TableName documents = TableName.valueOf("documents");
    HTableDescriptor desc = new HTableDescriptor(documents);
    HColumnDescriptor family = new HColumnDescriptor("c");
    family.setCompressionType(Algorithm.GZ);
    family.setBloomFilterType(BloomType.NONE);
    desc.addFamily(family);
    UniformSplit uniformSplit = new UniformSplit();
    admin.createTable(desc, uniformSplit.split(8));
    // end::CREATE[]
    LOG.info("Table successfuly created");
  }
}

Source File: HBaseBasedAuditRepository.java From incubator-atlas with Apache License 2.0

6 votes

private void createTableIfNotExists() throws AtlasException {
    Admin admin = null;
    try {
        admin = connection.getAdmin();
        LOG.info("Checking if table {} exists", tableName.getNameAsString());
        if (!admin.tableExists(tableName)) {
            LOG.info("Creating table {}", tableName.getNameAsString());
            HTableDescriptor tableDescriptor = new HTableDescriptor(tableName);
            HColumnDescriptor columnFamily = new HColumnDescriptor(COLUMN_FAMILY);
            columnFamily.setMaxVersions(1);
            columnFamily.setDataBlockEncoding(DataBlockEncoding.FAST_DIFF);
            columnFamily.setCompressionType(Compression.Algorithm.GZ);
            columnFamily.setBloomFilterType(BloomType.ROW);
            tableDescriptor.addFamily(columnFamily);
            admin.createTable(tableDescriptor);
        } else {
            LOG.info("Table {} exists", tableName.getNameAsString());
        }
    } catch (IOException e) {
        throw new AtlasException(e);
    } finally {
        close(admin);
    }
}

Source File: HBaseCreateTable.java From Kafka-Spark-Hbase-Example with Apache License 2.0

5 votes

public static void main(String[] args) throws IOException {
	if (args.length == 0) {
		System.out.println("CreateTable {tableName} {columnFamilyName}");
		return;
	}

	String tableName = args[0];
	String columnFamilyName = args[1];

	HBaseAdmin admin = new HBaseAdmin(new Configuration());

	HTableDescriptor tableDescriptor = new HTableDescriptor(); 
	tableDescriptor.setName(Bytes.toBytes(tableName));

	HColumnDescriptor columnDescriptor = new HColumnDescriptor(columnFamilyName);

	columnDescriptor.setCompressionType(Compression.Algorithm.SNAPPY);
	columnDescriptor.setBlocksize(64 * 1024);
	columnDescriptor.setBloomFilterType(BloomType.ROW);

	tableDescriptor.addFamily(columnDescriptor);

	//tableDescriptor.setValue(tableDescriptor.SPLIT_POLICY, ConstantSizeRegionSplitPolicy.class.getName());

	System.out.println("-Creating Table");
	admin.createTable(tableDescriptor);

	admin.close();
	System.out.println("-Done");
}

Source File: HbaseTableConnection.java From foxtrot with Apache License 2.0

5 votes

private HTableDescriptor constructHTableDescriptor(final Table table) {
    String tableName = TableUtil.getTableName(hbaseConfig, table);

    HTableDescriptor hTableDescriptor = new HTableDescriptor(TableName.valueOf(tableName));
    HColumnDescriptor hColumnDescriptor = new HColumnDescriptor(DEFAULT_FAMILY_NAME);
    hColumnDescriptor.setCompressionType(Compression.Algorithm.GZ);
    hColumnDescriptor.setTimeToLive(Math.toIntExact(TimeUnit.DAYS.toSeconds(table.getTtl())));
    hTableDescriptor.addFamily(hColumnDescriptor);
    return hTableDescriptor;
}

Source File: CreateTable.java From HBase-ToHDFS with Apache License 2.0

5 votes

private static void createTable(String tableName, String columnFamilyName,
    short regionCount, long regionMaxSize, HBaseAdmin admin)
    throws IOException {
  System.out.println("Creating Table: " + tableName);
  
  HTableDescriptor tableDescriptor = new HTableDescriptor(); 
  tableDescriptor.setName(Bytes.toBytes(tableName));
  
  HColumnDescriptor columnDescriptor = new HColumnDescriptor(columnFamilyName);
  
  columnDescriptor.setCompressionType(Compression.Algorithm.SNAPPY);
  columnDescriptor.setBlocksize(64 * 1024);
  columnDescriptor.setBloomFilterType(BloomType.ROW);
  
  tableDescriptor.addFamily(columnDescriptor);
  
  tableDescriptor.setMaxFileSize(regionMaxSize);
  tableDescriptor.setValue(tableDescriptor.SPLIT_POLICY, ConstantSizeRegionSplitPolicy.class.getName());
  
  tableDescriptor.setDeferredLogFlush(true);
  
  regionCount = (short)Math.abs(regionCount);
  
  int regionRange = Short.MAX_VALUE/regionCount;
  int counter = 0;
  
  byte[][] splitKeys = new byte[regionCount][];
  for (int i = 0 ; i < splitKeys.length; i++) {
    counter = counter + regionRange;
    String key = StringUtils.leftPad(Integer.toString(counter), 5, '0');
    splitKeys[i] = Bytes.toBytes(key); 
    System.out.println(" - Split: " + i + " '" + key + "'");
  }
  
  admin.createTable(tableDescriptor, splitKeys);
}

Source File: HBaseSITestEnv.java From spliceengine with GNU Affero General Public License v3.0

5 votes

public static HColumnDescriptor createDataFamily() {
    HColumnDescriptor snapshot = new HColumnDescriptor(SIConstants.DEFAULT_FAMILY_BYTES);
    snapshot.setMaxVersions(Integer.MAX_VALUE);
    snapshot.setCompressionType(Compression.Algorithm.NONE);
    snapshot.setInMemory(true);
    snapshot.setBlockCacheEnabled(true);
    snapshot.setBloomFilterType(BloomType.ROW);
    return snapshot;
}

Source File: App.java From hadoop-arch-book with Apache License 2.0

5 votes

private static boolean createTable(byte[] tableName, byte[] columnFamilyName,
    short regionCount, long regionMaxSize, HBaseAdmin admin)
    throws IOException {

  if (admin.tableExists(tableName)) {
    return false;
  }

  HTableDescriptor tableDescriptor = new HTableDescriptor();
  tableDescriptor.setName(tableName);

  HColumnDescriptor columnDescriptor = new HColumnDescriptor(columnFamilyName);

  columnDescriptor.setCompressionType(Compression.Algorithm.SNAPPY);
  columnDescriptor.setBlocksize(64 * 1024);
  columnDescriptor.setBloomFilterType(BloomType.ROW);
  columnDescriptor.setMaxVersions(10);
  tableDescriptor.addFamily(columnDescriptor);

  tableDescriptor.setMaxFileSize(regionMaxSize);
  tableDescriptor.setValue(tableDescriptor.SPLIT_POLICY,
      ConstantSizeRegionSplitPolicy.class.getName());

  tableDescriptor.setDeferredLogFlush(true);

  regionCount = (short) Math.abs(regionCount);

  int regionRange = Short.MAX_VALUE / regionCount;
  int counter = 0;

  byte[][] splitKeys = new byte[regionCount][];
  for (byte[] splitKey : splitKeys) {
    counter = counter + regionRange;
    String key = StringUtils.leftPad(Integer.toString(counter), 5, '0');
    splitKey = Bytes.toBytes(key);
    System.out.println(" - Split: " + splitKey);
  }
  return true;
}

Source File: HBaseCreateTable.java From SparkOnALog with Apache License 2.0

5 votes

public static void main(String[] args) throws IOException {
	if (args.length == 0) {
		System.out.println("CreateTable {tableName} {columnFamilyName}");
		return;
	}

	String tableName = args[0];
	String columnFamilyName = args[1];

	HBaseAdmin admin = new HBaseAdmin(new Configuration());

	HTableDescriptor tableDescriptor = new HTableDescriptor(); 
	tableDescriptor.setName(Bytes.toBytes(tableName));

	HColumnDescriptor columnDescriptor = new HColumnDescriptor(columnFamilyName);

	columnDescriptor.setCompressionType(Compression.Algorithm.SNAPPY);
	columnDescriptor.setBlocksize(64 * 1024);
	columnDescriptor.setBloomFilterType(BloomType.ROW);

	tableDescriptor.addFamily(columnDescriptor);

	//tableDescriptor.setValue(tableDescriptor.SPLIT_POLICY, ConstantSizeRegionSplitPolicy.class.getName());

	System.out.println("-Creating Table");
	admin.createTable(tableDescriptor);

	admin.close();
	System.out.println("-Done");
}

Source File: Configure.java From learning-hadoop with Apache License 2.0

5 votes

public static void configColumnFamily(HColumnDescriptor desc) {
  desc.setMaxVersions(1);
  // 设置使用的过滤器的类型---
  // setBloomFilter:指定是否使用BloomFilter,可提高随机查询效率。默认关闭
  desc.setBloomFilterType(BloomType.ROW);
  // 设定数据压缩类型。默认无压缩
  desc.setCompressionType(COMPRESS_TYPE);
}

Source File: CubeHTableUtil.java From kylin with Apache License 2.0

4 votes

public static HColumnDescriptor createColumnFamily(KylinConfig kylinConfig, String cfName, boolean isMemoryHungry) {
    HColumnDescriptor cf = new HColumnDescriptor(cfName);
    cf.setMaxVersions(1);

    if (isMemoryHungry) {
        cf.setBlocksize(kylinConfig.getHbaseDefaultBlockSize());
    } else {
        cf.setBlocksize(kylinConfig.getHbaseSmallFamilyBlockSize());
    }

    String hbaseDefaultCC = kylinConfig.getHbaseDefaultCompressionCodec().toLowerCase(Locale.ROOT);
    switch (hbaseDefaultCC) {
    case "snappy": {
        logger.info("hbase will use snappy to compress data");
        cf.setCompressionType(Algorithm.SNAPPY);
        break;
    }
    case "lzo": {
        logger.info("hbase will use lzo to compress data");
        cf.setCompressionType(Algorithm.LZO);
        break;
    }
    case "gz":
    case "gzip": {
        logger.info("hbase will use gzip to compress data");
        cf.setCompressionType(Algorithm.GZ);
        break;
    }
    case "lz4": {
        logger.info("hbase will use lz4 to compress data");
        cf.setCompressionType(Algorithm.LZ4);
        break;
    }
    case "none":
    default: {
        logger.info("hbase will not use any compression algorithm to compress data");
        cf.setCompressionType(Algorithm.NONE);
    }
    }

    try {
        String encodingStr = kylinConfig.getHbaseDefaultEncoding();
        DataBlockEncoding encoding = DataBlockEncoding.valueOf(encodingStr);
        cf.setDataBlockEncoding(encoding);
    } catch (Exception e) {
        logger.info("hbase will not use any encoding", e);
        cf.setDataBlockEncoding(DataBlockEncoding.NONE);
    }

    cf.setInMemory(false);
    cf.setBloomFilterType(BloomType.NONE);
    cf.setScope(kylinConfig.getHBaseReplicationScope());
    return cf;
}

Source File: HBaseCompat0_98.java From incubator-atlas with Apache License 2.0

4 votes

@Override
public void setCompression(HColumnDescriptor cd, String algo) {
    cd.setCompressionType(Compression.Algorithm.valueOf(algo));
}

Source File: HBaseCompat1_1.java From incubator-atlas with Apache License 2.0

4 votes

@Override
public void setCompression(HColumnDescriptor cd, String algo) {
    cd.setCompressionType(Compression.Algorithm.valueOf(algo));
}

Source File: CreateHTableJob.java From Kylin with Apache License 2.0

4 votes

@Override
public int run(String[] args) throws Exception {
    Options options = new Options();

    options.addOption(OPTION_CUBE_NAME);
    options.addOption(OPTION_PARTITION_FILE_PATH);
    options.addOption(OPTION_HTABLE_NAME);
    parseOptions(options, args);

    Path partitionFilePath = new Path(getOptionValue(OPTION_PARTITION_FILE_PATH));

    String cubeName = getOptionValue(OPTION_CUBE_NAME).toUpperCase();
    KylinConfig config = KylinConfig.getInstanceFromEnv();
    CubeManager cubeMgr = CubeManager.getInstance(config);
    CubeInstance cube = cubeMgr.getCube(cubeName);
    CubeDesc cubeDesc = cube.getDescriptor();

    String tableName = getOptionValue(OPTION_HTABLE_NAME).toUpperCase();
    HTableDescriptor tableDesc = new HTableDescriptor(TableName.valueOf(tableName));
    // https://hbase.apache.org/apidocs/org/apache/hadoop/hbase/regionserver/ConstantSizeRegionSplitPolicy.html
    tableDesc.setValue(HTableDescriptor.SPLIT_POLICY, ConstantSizeRegionSplitPolicy.class.getName());
    tableDesc.setValue(IRealizationConstants.HTableTag, config.getMetadataUrlPrefix());

    Configuration conf = HBaseConfiguration.create(getConf());
    HBaseAdmin admin = new HBaseAdmin(conf);

    try {
        if (User.isHBaseSecurityEnabled(conf)) {
            // add coprocessor for bulk load
            tableDesc.addCoprocessor("org.apache.hadoop.hbase.security.access.SecureBulkLoadEndpoint");
        }

        for (HBaseColumnFamilyDesc cfDesc : cubeDesc.getHBaseMapping().getColumnFamily()) {
            HColumnDescriptor cf = new HColumnDescriptor(cfDesc.getName());
            cf.setMaxVersions(1);

            if (LZOSupportnessChecker.getSupportness()) {
                logger.info("hbase will use lzo to compress data");
                cf.setCompressionType(Algorithm.LZO);
            } else {
                logger.info("hbase will not use lzo to compress data");
            }

            cf.setDataBlockEncoding(DataBlockEncoding.FAST_DIFF);
            cf.setInMemory(false);
            cf.setBlocksize(4 * 1024 * 1024); // set to 4MB
            tableDesc.addFamily(cf);
        }

        byte[][] splitKeys = getSplits(conf, partitionFilePath);

        if (admin.tableExists(tableName)) {
            // admin.disableTable(tableName);
            // admin.deleteTable(tableName);
            throw new RuntimeException("HBase table " + tableName + " exists!");
        }

        DeployCoprocessorCLI.deployCoprocessor(tableDesc);

        admin.createTable(tableDesc, splitKeys);
        logger.info("create hbase table " + tableName + " done.");

        return 0;
    } catch (Exception e) {
        printUsage(options);
        e.printStackTrace(System.err);
        logger.error(e.getLocalizedMessage(), e);
        return 2;
    } finally {
        admin.close();
    }
}

Source File: HBaseCompat1_0.java From incubator-atlas with Apache License 2.0

4 votes

@Override
public void setCompression(HColumnDescriptor cd, String algo) {
    cd.setCompressionType(Compression.Algorithm.valueOf(algo));
}

Source File: HBaseCompat1_1.java From titan1withtp3.1 with Apache License 2.0

4 votes

@Override
public void setCompression(HColumnDescriptor cd, String algo) {
    cd.setCompressionType(Compression.Algorithm.valueOf(algo));
}

Source File: HBaseCompat1_0.java From titan1withtp3.1 with Apache License 2.0

4 votes

@Override
public void setCompression(HColumnDescriptor cd, String algo) {
    cd.setCompressionType(Compression.Algorithm.valueOf(algo));
}

Source File: HBaseSchemaManager.java From replicator with Apache License 2.0

4 votes

public synchronized void createHBaseTableIfNotExists(String hbaseTableName) throws IOException {

        if (!DRY_RUN) {
            hbaseTableName = hbaseTableName.toLowerCase();
            try ( Admin admin = connection.getAdmin()) {

                if (seenHBaseTables.containsKey(hbaseTableName)) {
                    return;
                }

                if (connection == null) {
                    connection = ConnectionFactory.createConnection(storageConfig.getConfig());
                }

                TableName tableName;

                String namespace = (String) configuration.get(HBaseApplier.Configuration.TARGET_NAMESPACE);
                if (namespace.isEmpty()) {
                    tableName = TableName.valueOf(hbaseTableName);
                } else {
                    tableName = TableName.valueOf(namespace, hbaseTableName);
                }

                if (admin.tableExists(tableName)) {
                    LOG.warn("Table " + tableName + " exists in HBase, but not in schema cache. Probably a case of a table that was dropped and than created again");
                    seenHBaseTables.put(hbaseTableName, 1);
                } else {
                    HTableDescriptor tableDescriptor = new HTableDescriptor(tableName);
                    HColumnDescriptor cd = new HColumnDescriptor("d");

                    if (USE_SNAPPY) {
                        cd.setCompressionType(Compression.Algorithm.SNAPPY);
                    }

                    cd.setMaxVersions(MIRRORED_TABLE_NUMBER_OF_VERSIONS);
                    tableDescriptor.addFamily(cd);
                    tableDescriptor.setCompactionEnabled(true);

                    admin.createTable(tableDescriptor);

                    seenHBaseTables.put(hbaseTableName, 1);

                    LOG.warn("Created hbase table " + hbaseTableName);
                }

            } catch (IOException e) {
                throw new IOException("Failed to create table in HBase", e);
            }
        }
    }

Source File: TableCommand.java From pinpoint with Apache License 2.0

4 votes

private HColumnDescriptor newColumnDescriptor(ColumnFamilyChange columnFamilyChange) {
    HColumnDescriptor hcd = new HColumnDescriptor(columnFamilyChange.getName());
    ColumnFamilyConfiguration columnFamilyConfiguration = columnFamilyChange.getColumnFamilyConfiguration();
    Boolean blockCacheEnabled = columnFamilyConfiguration.getBlockCacheEnabled();
    if (blockCacheEnabled != null) {
        hcd.setBlockCacheEnabled(blockCacheEnabled);
    }
    Integer replicationScope = columnFamilyConfiguration.getReplicationScope();
    if (replicationScope != null) {
        hcd.setScope(replicationScope);
    }
    Boolean inMemory = columnFamilyConfiguration.getInMemory();
    if (inMemory != null) {
        hcd.setInMemory(inMemory);
    }
    Integer timeToLive = columnFamilyConfiguration.getTimeToLive();
    if (timeToLive != null) {
        hcd.setTimeToLive(timeToLive);
    }
    ColumnFamilyConfiguration.DataBlockEncoding dataBlockEncoding =
            columnFamilyConfiguration.getDataBlockEncoding();
    if (dataBlockEncoding != null) {
        hcd.setDataBlockEncoding(DataBlockEncoding.valueOf(dataBlockEncoding.name()));
    }
    Integer blockSize = columnFamilyConfiguration.getBlockSize();
    if (blockSize != null) {
        hcd.setBlocksize(blockSize);
    }
    Integer maxVersions = columnFamilyConfiguration.getMaxVersions();
    if (maxVersions != null) {
        hcd.setMaxVersions(maxVersions);
    }
    Integer minVersions = columnFamilyConfiguration.getMinVersions();
    if (minVersions != null) {
        hcd.setMinVersions(minVersions);
    }
    ColumnFamilyConfiguration.BloomFilter bloomFilter = columnFamilyConfiguration.getBloomFilter();
    if (bloomFilter != null) {
        hcd.setBloomFilterType(BloomType.valueOf(bloomFilter.name()));
    }
    if (compressionAlgorithm != Compression.Algorithm.NONE) {
        hcd.setCompressionType(compressionAlgorithm);
    }
    return hcd;
}

Source File: CubeHTableUtil.java From kylin-on-parquet-v2 with Apache License 2.0

4 votes

public static HColumnDescriptor createColumnFamily(KylinConfig kylinConfig, String cfName, boolean isMemoryHungry) {
    HColumnDescriptor cf = new HColumnDescriptor(cfName);
    cf.setMaxVersions(1);

    if (isMemoryHungry) {
        cf.setBlocksize(kylinConfig.getHbaseDefaultBlockSize());
    } else {
        cf.setBlocksize(kylinConfig.getHbaseSmallFamilyBlockSize());
    }

    String hbaseDefaultCC = kylinConfig.getHbaseDefaultCompressionCodec().toLowerCase(Locale.ROOT);
    switch (hbaseDefaultCC) {
    case "snappy": {
        logger.info("hbase will use snappy to compress data");
        cf.setCompressionType(Algorithm.SNAPPY);
        break;
    }
    case "lzo": {
        logger.info("hbase will use lzo to compress data");
        cf.setCompressionType(Algorithm.LZO);
        break;
    }
    case "gz":
    case "gzip": {
        logger.info("hbase will use gzip to compress data");
        cf.setCompressionType(Algorithm.GZ);
        break;
    }
    case "lz4": {
        logger.info("hbase will use lz4 to compress data");
        cf.setCompressionType(Algorithm.LZ4);
        break;
    }
    case "none":
    default: {
        logger.info("hbase will not use any compression algorithm to compress data");
        cf.setCompressionType(Algorithm.NONE);
    }
    }

    try {
        String encodingStr = kylinConfig.getHbaseDefaultEncoding();
        DataBlockEncoding encoding = DataBlockEncoding.valueOf(encodingStr);
        cf.setDataBlockEncoding(encoding);
    } catch (Exception e) {
        logger.info("hbase will not use any encoding", e);
        cf.setDataBlockEncoding(DataBlockEncoding.NONE);
    }

    cf.setInMemory(false);
    cf.setBloomFilterType(BloomType.NONE);
    cf.setScope(kylinConfig.getHBaseReplicationScope());
    return cf;
}

Java Code Examples for org.apache.hadoop.hbase.HColumnDescriptor#setCompressionType()