All Products
Search

This Product

    Data Lake Formation:Manage Format tables

    Document Center

    Data Lake Formation:Manage Format tables

    Last Updated:Dec 05, 2025

    This topic describes the basic operations on Format tables in Data Lake Formation.

    Introduction

    • Data management: DLF manages metadata and data files.

    • Storage system: DLF automatically generates a data storage path based on a UUID, so you do not need to specify one.

    • Delete behavior: Data retained for 1 day by default, then permanently deleted.

    Create a table

    1. Log on to the DLF console.

    2. In the left navigation menu, select Catalogs, and click your catalog name.

    3. In the Database section, click your database name.

    4. Click Create Table.

    5. Configure the table as shown below and click OK.

      Configuration Item

      Description

      Table Format

      Select Format Table.

      Table Name

      Enter a name, which must be unique within the database.

      Table Description

      Enter a description.

      Columns

      Define the columns in the table.

      User-defined Table Properties

      file.format: Specifies the file format. Supported formats are Parquet, ORC, CSV, and JSON.

      You can add custom properties. During table creation, these properties overwrite the default properties. For a list of applicable configuration items, see the Configuration in the Apache Paimon documentation.

      SQL examples

      A Format table is a Paimon table managed by DLF. DLF automatically generates a storage path based on a UUID and performs table optimization and maintenance in the background. This lets you focus on your business logic.

      • Flink SQL

        CREATE TABLE my_csv_table (
    a INT,
    b STRING
) WITH (
    'type'='format-table',
    'file.format'='csv'
);

      • Spark SQL

        CREATE TABLE my_csv_table (
    a INT,
    b STRING
) USING CSV

      Advanced configuration

      1. Support value-only partition paths

        By default, partition paths use the key=value format, such as year=2025. If your storage directory contains only the partition value, such as table-path/2025/xxxx.csv, you must explicitly enable this option.

        Set the parameter: format-table.partition-path-only-value: true.

      2. Set file compression

        You can use the format-table.file.compression parameter to specify the compression algorithm for output files. If this parameter is not specified, the system uses the following default values:

        File type

        Default

        parquet

        snappy

        avro/orc

        zstd

        csv/json

        none (no compression)

      View a table

      1. In the Database section, click your database name.

      2. On the Tables tab, click your table name.

      3. On the Table Details tab, view the table's basic information and columns.

      4. On the Permissions tab, grant table permissions to DLF users or roles. For more information, see Data Authorization Management.

      Delete a table

      Warning

      Deleted table data is retained for 1 day to prevent accidental deletion. Data is permanently deleted afterward.

      1. In the Database section, click your database name.

      2. On the Tables tab, click Delete in the Actions column of your target table.

      3. In the dialog box, click OK to confirm deletion.