面试题与答案

Sqoop is a tool designed for efficiently transferring bulk data between Apache Hadoop and structured datastores such as relational databases.

The --target-dir option specifies the HDFS directory where the imported data will be stored.

The --warehouse-dir option specifies the base directory in HDFS where imported data is stored.

The --update-key option specifies the column(s) used to identify rows for updates when performing an export.

The --as-textfile option in Sqoop specifies that the data should be stored in text format in HDFS during import.

A full load import in Sqoop can be done by using the --m (or --num-mappers) option with a value of 1 to import the data using a single mapper.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --m 1

These options are used to specify the representation of NULL values in the imported data for string and non-string columns, respectively.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --null-string --null-non-string -1

The --columns option allows you to specify a comma-separated list of columns to import, excluding others from the source table.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --columns id,name

The --fetch-size option specifies the number of rows to fetch in each round trip between Sqoop and the database during import.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --fetch-size 100

The import command in Sqoop is used to import data from a relational database into Hadoop.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --target-dir /user/hadoop/mytable

Incremental imports in Sqoop can be done using the --incremental option. You need to specify the mode and the column to use for tracking changes.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --incremental append --check-column id --last-value 100

The export command in Sqoop is used to export data from Hadoop to a relational database.

Example:

sqoop export --connect jdbc:mysql://localhost:3306/db --table mytable --export-dir /user/hadoop/mytable

The metastore in Sqoop is a central repository that stores metadata related to Sqoop jobs, such as saved jobs, connection information, and job history.

The --map-column-java option allows you to specify how the columns from the database table should be mapped to Java types during import.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --map-column-java id=String,value=Double

In free-form query import, you can specify a SQL query to extract data, while in table-based import, you directly import an entire table.

The --merge-key option is used during the Sqoop merge operation to specify the columns used for identifying rows to merge.

The --query option allows you to specify a SQL SELECT statement to retrieve data during Sqoop import.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --query SELECT * FROM mytable WHERE $CONDITIONS --split-by id

The --boundary-query option allows you to specify a SQL query that is used to determine the range of values for the splitting column.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --boundary-query SELECT MIN(id), MAX(id) FROM mytable

The --boundary-query option allows you to specify a SQL query that is used to determine the range of values for the splitting column.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --boundary-query SELECT MIN(id), MAX(id) FROM mytable

You can import data into Hive using Sqoop by specifying the --hive-import option along with the target Hive table using --hive-table.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --hive-import --hive-table myhivetable

The --hive-overwrite option in Sqoop is used to overwrite existing data in the Hive table during import.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --hive-import --hive-table myhivetable --hive-overwrite

To update existing records during export in Sqoop, you can use the --update-key option along with --update-mode option set to allowinsert.

Example:

sqoop export --connect jdbc:mysql://localhost:3306/db --table mytable --update-key id --update-mode allowinsert --export-dir /user/hadoop/mytable

The --hcatalog-database option specifies the HCatalog database name when importing data into HCatalog.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --hcatalog-import --hcatalog-database mydatabase

The --direct option enables direct export or import between Hadoop and the database without using HDFS as an intermediary.

The --direct-split-size option is used to specify the number of bytes per split when using direct mode for imports and exports.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --direct --direct-split-size 1000000

The --direct-import option is used to import data directly into the database without using HDFS as an intermediate storage.

The --validate option is used to perform data validation during import by comparing the source and target data counts.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --validate

The --direct-split-size option is used to specify the number of bytes per split when using direct mode for imports and exports.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --direct --direct-split-size 1000000

The --direct-split-size option is used to specify the number of bytes per split when using direct mode for imports and exports.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --direct --direct-split-size 1000000

The --autoreset-to-one-mapper option automatically resets the number of mappers to one if the initial split size is larger than the total number of rows in the table.

Example:

sqoop import --connect jdbc:mysql://localhost:3306/db --table mytable --autoreset-to-one-mapper