hawq register
Loads and registers AO or Parquet-formatted tables in HDFS into a corresponding table in HAWQ.
Synopsis
Usage 1:
hawq register [<connection_options>] [-f <hdfsfilepath>] [-e <Eof>] <tablename>
Usage 2:
hawq register [<connection_options>] [-c <configfilepath>][-F] <tablename>
Connection Options:
[-h | --host <hostname>]
[-p | --port <port>]
[-U | --user <username>]
[-d | --database <database>]
Misc. Options:
[-f | --filepath <filepath>]
[-e | --eof<eof>]
[-F | --force ]
[-c | --config <yml_config>]
hawq register help | -?
hawq register --version
Prerequisites
The client machine where hawq register
is executed must meet the following conditions:
- All hosts in your HAWQ cluster (master and segments) must have network access between them and the hosts containing the data to be loaded.
- The Hadoop client must be configured and the hdfs filepath specified.
- The files to be registered and the HAWQ table must be located in the same HDFS cluster.
- The target table DDL is configured with the correct data type mapping.
Description
hawq register
is a utility that loads and registers existing data files or folders in HDFS into HAWQ internal tables, allowing HAWQ to directly read the data and use internal table processing for operations such as transactions and high performance, without needing to load or copy it. Data from the file or directory specified by <hdfsfilepath> is loaded into the appropriate HAWQ table directory in HDFS and the utility updates the corresponding HAWQ metadata for the files.
You can use hawq register
to:
- Load and register external Parquet-formatted file data generated by an external system such as Hive or Spark.
- Recover cluster data from a backup cluster.
Two usage models are available.
Usage Model 1: Register file data to an existing table.
hawq register [-h hostname] [-p port] [-U username] [-d databasename] [-f filepath] [-e eof]<tablename>
Metadata for the Parquet file(s) and the destination table must be consistent. Different data types are used by HAWQ tables and Parquet files, so the data is mapped. Refer to the section Data Type Mapping below. You must verify that the structure of the Parquet files and the HAWQ table are compatible before running hawq register
.
Limitations
Only HAWQ or Hive-generated Parquet tables are supported. Hash tables and partitioned tables are not supported in this use model.
Usage Model 2: Use information from a YAML configuration file to register data
hawq register [-h hostname] [-p port] [-U username] [-d databasename] [-c configfile] [--force] <tablename>
Files generated by the hawq extract
command are registered through use of metadata in a YAML configuration file. Both AO and Parquet tables can be registered. Tables need not exist in HAWQ before being registered.
The register process behaves differently, according to different conditions.
- Existing tables have files appended to the existing HAWQ table.
- If a table does not exist, it is created and registered into HAWQ.
- If the --force option is used, the data in existing catalog tables is erased and re-registered.
Limitations for Registering Hive Tables to HAWQ
The currently-supported data types for generating Hive tables into HAWQ tables are: boolean, int, smallint, tinyint, bigint, float, double, string, binary, char, and varchar.
The following HIVE data types cannot be converted to HAWQ equivalents: timestamp, decimal, array, struct, map, and union.
Only single-level partitioned tables are supported.
Data Type Mapping
HAWQ and Parquet tables and HIVE and HAWQ tables use different data types. Mapping must be used for compatibility. You are responsible for making sure your implementation is mapped to the appropriate data type before running hawq register
. The tables below show equivalent data types, if available.
Table 1. HAWQ to Parquet Mapping
HAWQ Data Type | Parquet Data Type |
---|---|
bool | boolean |
int2/int4/date | int32 |
int8/money | int64 |
time/timestamptz/timestamp | int64 |
float4 | float |
float8 | double |
bit/varbit/bytea/numeric | Byte array |
char/bpchar/varchar/name | Byte array |
text/xml/interval/timetz | Byte array |
macaddr/inet/cidr | Byte array |
Additional HAWQ-to-Parquet Mapping
point:
group {
required int x;
required int y;
}
circle:
group {
required int x;
required int y;
required int r;
}
box:
group {
required int x1;
required int y1;
required int x2;
required int y2;
}
iseg:
group {
required int x1;
required int y1;
required int x2;
required int y2;
}
path:
group {
repeated group {
required int x;
required int y;
}
}
Table 2. HIVE to HAWQ Mapping
HIVE Data Type | HAWQ Data Type |
---|---|
boolean | bool |
tinyint | int2 |
smallint | int2/smallint |
int | int4 / int |
bigint | int8 / bigint |
float | float4 |
double | float8 |
string | varchar |
binary | bytea |
char | char |
varchar | varchar |
Options
General Options
Connection Options
$PGHOST
or defaults to localhost
.$PGPORT
or defaults to 5432.$PGUSER
or defaults to the current system user name.postgres
Miscellaneous Options
The following options are used with specific use models.
pg_aoseg.pg_paqseg_$relid
and re-registers files to a specified table. The HDFS files are not removed or modified. To use this option for recovery, data is assumed to be periodically imported to the cluster to be recovered. Used with Usage Model 2.Example: Usage Model 2
This example shows how to register files using a YAML configuration file. This file is usually generated by the hawq extract
command.
Create a table and insert data into the table:
=> CREATE TABLE paq1(a int, b varchar(10))with(appendonly=true, orientation=parquet);`
=> INSERT INTO paq1 values(generate_series(1,1000), 'abcde');
Extract the table’s metadata.
hawq extract -o paq1.yml paq1
Use the YAML file to register the new table paq2:
hawq register --config paq1.yml paq2
Select the new table to determine if the content has already been registered:
=> SELECT count(*) FROM paq2;
The result should return 1000.