hbase，根据前缀匹配进行搜索并分批次获取结果

1,获取hbase client

/**
   * Create a new Connection instance using the passed <code>conf</code> instance. Connection
   * encapsulates all housekeeping for a connection to the cluster. All tables and interfaces
   * created from returned connection share zookeeper connection, meta cache, and connections
   * to region servers and masters.
   * <br>
   * The caller is responsible for calling {@link Connection#close()} on the returned
   * connection instance.
   *
   * Typical usage:
   * <pre>
   * Connection connection = ConnectionFactory.createConnection(conf);
   * Table table = connection.getTable(TableName.valueOf("mytable"));
   * try {
   *   table.get(...);
   *   ...
   * } finally {
   *   table.close();
   *   connection.close();
   * }
   * </pre>
   *
   * @param conf configuration
   * @return Connection object for <code>conf</code>
   */
  public static Connection createConnection(Configuration conf) throws IOException {
    return createConnection(conf, null, null);
  }

　　

2,根据前缀匹配进行搜索并分批次获取结果

 private static AtomicLong filterAndDeleRowkeysByYearMonth(String yearMonth,String tableName,String devNo,String[] dayDateRange,String zkIp,String esIp,String esTableName) throws IOException {
//        Set dates = new HashSet<>(Arrays.asList("20180124")); //需要查询的日期集合
        Set dates = new HashSet<>(Arrays.asList(dayDateRange));
        Table table = HBaseUtil.getTable(tableName, zkIp);
        TransportClient client = ESConnectFactory.getTransportClientInstance(esIp);
        Scan s = new Scan();
        s.setCaching(10000000);
        s.setFilter(new PrefixFilter(yearMonth.getBytes()));
//        s.setFilter(new PrefixFilter(yearMonth.getBytes()));
//        s.setMaxResultSize(100000000);
        ResultScanner resultScanner = table.getScanner(s);
        AtomicLong atomicLong1 = new AtomicLong();
        Result[] next = null;
        do {
            next = resultScanner.next(100000);
//            long l = runDelete(next, devNo, dates, atomicLong, table, client);
            ThreadPoolManager.getInstance().getProduceThreadPool().execute(new RowKeyProducerThread(next, devNo, dates));
            ThreadPoolManager.getInstance().getConsumeThreadPool().execute(new RowKeyConsumerThread(table,devNo,atomicLong1,client, esTableName));
        }while(next != null && next.length > 0 );
        try {
            Thread.sleep(10000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        ThreadPoolManager.getInstance().getProduceThreadPool().shutdown();
        ThreadPoolManager.getInstance().getConsumeThreadPool().shutdown();
        return atomicLong1;
    }

　　

转载于:https://www.cnblogs.com/yanyuechao/p/8367517.html