AnalyticDB:Use external tables to import data to Data Warehouse Edition

Prerequisites

• An AnalyticDB for MySQL cluster of V3.1.4.4 or later is created.

• An HDFS data file is in the CSV, Parquet, or ORC format.

• An HDFS cluster is created, and the data that you want to import is stored in an HDFS folder. In this example, the data is stored in the hdfs_import_test_data.csv folder.

• The following service access ports for the AnalyticDB for MySQL cluster are configured in the HDFS cluster:

  • namenode: used to read and write metadata of a file system. You can configure the port number by using the fs.defaultFS parameter. The default port number is 8020.

    For information about the detailed configurations, see core-default.xml.

  • datanode: used to read and write data. You can configure the port number by using the dfs.datanode.address parameter. The default port number is 50010.

    For information about the detailed configurations, see hdfs-default.xml.

• If the AnalyticDB for MySQL cluster is in elastic mode, you must turn on ENI in the Network Information section of the Cluster Information page.

Procedure

1. Connect to the AnalyticDB for MySQL cluster. For more information, see Connect to an AnalyticDB for MySQL cluster.

2. Create a database. For more information, see Create a database.

   In this example, a database named adb_demo is created in the AnalyticDB for MySQL cluster.

3. Create an external table in the CSV, Parquet, or ORC format in the destination database adb_demo by executing a CREATE TABLE statement.

   • To create a standard external table, see the "Create an HDFS external table" section of this topic.

   • To create a partitioned external table, see the "Create a partitioned HDFS external table" section of this topic.

4. Create a destination table.

   You can execute one of the following statements to create a destination table in the destination database adb_demo to store data imported from HDFS.

   • The following statement shows how to create a destination table corresponding to the standard external table. In this example, the destination table is named adb_hdfs_import_test.

     CREATE TABLE IF NOT EXISTS adb_hdfs_import_test
     (
         uid string,
         other string
     )
     DISTRIBUTED BY HASH(uid);

   • The following statement shows how to create a destination table corresponding to the partitioned external table. In this example, the destination table is named adb_hdfs_import_parquet_partition. To create the table, you must define ordinary columns (such as uid and other) and partition columns (such as p1, p2, and p3) in the statement.

     CREATE TABLE IF NOT EXISTS adb_hdfs_import_parquet_partition
     (
         uid string,
         other string,
         p1 date,
         p2 int,
         p3 varchar
     )
     DISTRIBUTED BY HASH(uid);

5. Import data from HDFS to the destination AnalyticDB for MySQL cluster.

   You can use one of the following methods to import data. The syntax to import data by using a partitioned table is the same as the syntax to import data by using a standard table.

   • Method 1 (recommended): Execute an INSERT OVERWRITE statement to import data. This method allows you to batch import data and provides good performance. If the import succeeds, the data is available for query. Otherwise, the data is rolled back to its previous state. Example:

     INSERT OVERWRITE adb_hdfs_import_test
     SELECT * FROM hdfs_import_test_external_table;

   • Method 2: Execute an INSERT INTO statement to import data. This method allows you to query imported data in real time. This method is suitable when you want to import a small amount of data. Example:

     INSERT INTO adb_hdfs_import_test
     SELECT * FROM hdfs_import_test_external_table;

   • Method 3: Submit an asynchronous task to import data. Example:

     SUBMIT JOB INSERT OVERWRITE adb_hdfs_import_test
     SELECT * FROM hdfs_import_test_external_table;

     The following information is returned:

     +---------------------------------------+
     | job_id                                |
     +---------------------------------------+
     | 2020112122202917203100908203303****** |
     +---------------------------------------+

     You can check the state of the asynchronous task based on the job_id value. For more information, see Asynchronously submit an import task.

What to do next

SELECT * FROM adb_hdfs_import_test LIMIT 100;

Create an HDFS external table

• Create an external table in the CSV format

  The following statement shows how to create an external table in the CSV format:

  CREATE TABLE IF NOT EXISTS hdfs_import_test_external_table
  (
      uid string,
      other string
  )
  ENGINE='HDFS'
  TABLE_PROPERTIES='{
      "format":"csv",
      "delimiter":",",
      "hdfs_url":"hdfs://172.17.***.***:9000/adb/hdfs_import_test_csv_data/hdfs_import_test_data.csv"
  }';

  Parameter	Required	Description
  ENGINE='HDFS'	Yes	The storage engine used for the external table. In this example, HDFS is used as the storage engine.
  TABLE_PROPERTIES		The connection information used by AnalyticDB for MySQL to access HDFS data.
  format		The format of the data file. When you create an external table in the CSV format, you must set this parameter to csv.
  delimiter		The column delimiter of the CSV file. In this example, a comma (,) is used as the delimiter.
  hdfs_url		The absolute URL of the destination data file or folder in the HDFS cluster. The URL must start with hdfs://. Example: hdfs://172.17.***.***:9000/adb/hdfs_import_test_csv_data/hdfs_import_test_data.csv
  partition_column	No	The partition column of the external table. Separate multiple columns with commas (,). For more information about the methods to define a partition column, see the "Create a partitioned HDFS external table" section of this topic.
  compress_type		The compression type of the data file. Only the GZIP compression type is supported for CSV files.
  skip_header_line_count		The number of header rows to skip when you import data. The first row of a CSV file is the table header. If you set this parameter to 1, the first row of the file is skipped when you import data. The default value of this parameter is 0, which indicates that no rows are skipped.
  hdfs_ha_host_port		If the high availability (HA) feature is configured for the HDFS cluster, you must specify the hdfs_ha_host_port parameter in the ip1:port1,ip2:port2 format when you create an external table. The IP addresses and port numbers of the primary and secondary NameNodes are specified in this parameter. Example: 192.168.xx.xx:8020,192.168.xx.xx:8021

• Create an external table in the Parquet or ORC format

  The following statement shows how to create an external table in the Parquet format:

  CREATE TABLE IF NOT EXISTS hdfs_import_test_external_table
  (
      uid string,
      other string
  )
  ENGINE='HDFS'
  TABLE_PROPERTIES='{
      "format":"parquet",
      "hdfs_url":"hdfs://172.17.***.***:9000/adb/hdfs_import_test_parquet_data/"
  }';

  Parameter	Required	Description
  ENGINE='HDFS'	Yes	The storage engine used for the external table. In this example, HDFS is used as the storage engine.
  TABLE_PROPERTIES		The connection information used by AnalyticDB for MySQL to access HDFS data.
  format		The format of the data file. When you create an external table in the Parquet format, you must set this parameter to parquet. When you create an external table in the ORC format, you must set this parameter to orc.
  hdfs_url		The absolute URL of the destination data file or folder in the HDFS cluster. The URL must start with hdfs://.
  partition_column	No	The partition column of the table. Separate multiple columns with commas (,). For more information about the methods to define a partition column, see the "Create a partitioned HDFS external table" section of this topic.
  hdfs_ha_host_port		If the high availability (HA) feature is configured for the HDFS cluster, you must specify the hdfs_ha_host_port parameter in the ip1:port1,ip2:port2 format when you create an external table. The IP addresses and port numbers of the primary and secondary NameNodes are specified in this parameter. Example: 192.168.xx.xx:8020,192.168.xx.xx:8021

  Note

  • The column names used in the statement to create an external table must be the same as those in the Parquet or ORC file. Column names are case-insensitive. The sequence of the columns in the statement must be the same as that in the Parquet or ORC file.

  • When you create an external table, you can choose only specific columns in a Parquet or ORC file as the columns of the external table. Columns that are not selected in the Parquet or ORC file are not imported.

  • If the statement used to create an external table contains a column that is not in the Parquet or ORC file, NULL is returned for this column.

  Data type mappings between Parquet and AnalyticDB for MySQL

  Basic type in Parquet	Logical type in Parquet	Data type in AnalyticDB for MySQL
  BOOLEAN	N/A	BOOLEAN
  INT32	INT_8	TINYINT
  INT32	INT_16	SMALLINT
  INT32	N/A	INT or INTEGER
  INT64	N/A	BIGINT
  FLOAT	N/A	FLOAT
  DOUBLE	N/A	DOUBLE
  FIXED_LEN_BYTE_ARRAY BINARY INT64 INT32	DECIMAL	DECIMAL
  BINARY	UTF-8	VARCHAR STRING JSON (available if the Parquet object contains a column of the JSON type)
  INT32	DATE	DATE
  INT64	TIMESTAMP_MILLIS	TIMESTAMP or DATETIME
  INT96	N/A	TIMESTAMP or DATETIME

  Important

  Parquet external tables that use columns of the STRUCT type cannot be created.

  Data type mappings between ORC and AnalyticDB for MySQL

  Data type in ORC	Data type in AnalyticDB for MySQL
  BOOLEAN	BOOLEAN
  BYTE	TINYINT
  SHORT	SMALLINT
  INT	INT or INTEGER
  LONG	BIGINT
  DECIMAL	DECIMAL
  FLOAT	FLOAT
  DOUBLE	DOUBLE
  BINARY STRING VARCHAR	VARCHAR STRING JSON (available if the ORC object contains a column of the JSON type)
  TIMESTAMP	TIMESTAMP or DATETIME
  DATE	DATE

  Important

  ORC external tables that use the LIST, STRUCT, or UNION type cannot be created. ORC external tables that use the MAP type cab be created but cannot be queried.

Create a partitioned HDFS external table

parquet_partition_classic/
├── p1=2020-01-01
│   ├── p2=4
│   │   ├── p3=SHANGHAI
│   │   │   ├── 000000_0
│   │   │   └── 000000_1
│   │   └── p3=SHENZHEN
│   │       └── 000000_0
│   └── p2=6
│       └── p3=SHENZHEN
│           └── 000000_0
├── p1=2020-01-02
│   └── p2=8
│       ├── p3=SHANGHAI
│       │   └── 000000_0
│       └── p3=SHENZHEN
│           └── 000000_0
└── p1=2020-01-03
    └── p2=6
        ├── p2=HANGZHOU
        └── p3=SHENZHEN
            └── 000000_0

CREATE TABLE IF NOT EXISTS hdfs_parquet_partition_table
(
  uid varchar,
  other varchar,
  p1 date,
  p2 int,
  p3 varchar
)
ENGINE='HDFS'
TABLE_PROPERTIES='{
  "hdfs_url":"hdfs://172.17.***.**:9000/adb/parquet_partition_classic/",
  "format":"parquet",  //Specify the format of the file. To create a CSV or ORC file, change the format value to csv or orc. 
  "partition_column":"p1, p2, p3"  //If you want to query the HDFS data by partition, you must specify the partition_column parameter in the statement used to create an external table when you import data to AnalyticDB for MySQL. 
}';