All Products
Search

This Product

    Dataphin:Development platform settings

    Document Center

    Dataphin:Development platform settings

    Last Updated:Jun 05, 2026

    Manage locks and ANALYZE command concurrency during development. Configure edit locks, object commits, query acceleration, and storage size updates.

    Limitations

    • You can use the Storage size update settings feature only when the compute engine is E-MapReduce 3.x, E-MapReduce 5.x, CDH 5.x, CDH 6.x, FusionInsight 8.x, Cloudera Data Platform 7.x, AsiaInfo DP 5.3, ArgoDB, Transwarp TDH 6.x, StarRocks, SelectDB, or Doris.

    • The SelectDB and Doris compute engines do not support Table management settings or Default compute engine for standard modeling.

    Permissions

    Requires a custom role with the Manage development platform settings permission, or the Super Admin or System Admin role.

    Access the development platform

    1. In the top navigation bar of the Dataphin homepage, choose Management Center > System Settings.

    2. In the left-side navigation pane, choose Platform Settings > Development Platform.

    Edit lock

    1. In the Edit Lock section, click the edit icon, enable Exclusive Edit Lock, and configure the lock.

      Parameter

      Description

      Exclusive Edit Lock

      When disabled, users can override each other's locks. When enabled, a locked object cannot be edited by others until the lock is released or expires.

      Lock Duration

      The lock expires after the specified inactivity period. Default: 30 minutes. Range: 5 minutes to 120 minutes.

      Automatically release when object is closed

      Releases the lock when the object's edit tab is closed.

      Automatically release on successful commit

      Releases the lock after a successful commit. Not released if the commit fails.

    2. Click OK.

      Click Restore Defaults to reset.

    Storage size update settings

    Hive does not automatically update storage sizes for tables written to HDFS by integration or real-time tasks, so the Data Catalog may not reflect current sizes. Dataphin can run the ANALYZE command automatically after task completion to fetch the latest storage data.

    1. In the Storage Size Update Settings section, click the edit icon, enable automatic storage size update, and configure concurrent connections.

      • Automatic Storage Size Update: Disabled by default. When enabled, Dataphin runs the ANALYZE command on the target Hive table after each successful task to update storage sizes. Increase concurrent connections to reduce execution time if your Hive server has sufficient capacity. High concurrency consumes significant compute resources and may affect other tasks.

      • Maximum Connections: Maximum concurrent connections for ANALYZE commands. Default: 5. Range: 1–200.

        Important

        ANALYZE commands running or pending for more than 24 hours are automatically terminated.

    2. Click OK.

      Note

      • When you enable Automatic Storage Size Update and click OK, the concurrent connection setting takes effect immediately. High concurrency consumes significant compute resources and may affect running tasks.

      • When you disable Automatic Storage Size Update, running or pending ANALYZE commands continue. Subsequent tasks will not trigger automatic updates. Run the ANALYZE command manually in Hive to update storage sizes.

    Node task settings

    1. In the Node task settings area, click Edit to configure default scheduling time and commit rules.

      • Creation

        Parameter

        Description

        Default priority

        The default priority for new integration, compute, and logical table tasks. Options: Lowest, Low, or Medium (default).

        Default scheduling time

        Randomly within an interval

        The default time interval is 00:00–03:00, and the default random interval is 5 minutes.

        • The end time must be later than the start time. Format: hh:mm (00:00–23:59).

        • The random time interval must be an integer from 1 to 30.

        Fixed time

        The default fixed time is 00:00.

        Format: hh:mm (00:00–23:59).

        Default Python version

        Default Python version for compute tasks, offline templates, and third-party packages. Options: Python 2.7, Python 3.7 (default), or Python 3.11.
        Note

        • By default, Default Scheduling Time is set to Randomly within an interval. You can change it to Fixed time as needed.

        • When you create an offline task (integration, compute, or logical table), the scheduling time is set based on this configuration.

          • If Default Scheduling Time is set to Randomly within an interval, a random time is selected based on the configured rules.

          • If Default Scheduling Time is set to Fixed time, the configured time is used.

      • Execution

        Hide logview URL if SQL contains account and password global variables: Prevents account and password global variables in SQL from appearing in plaintext in MaxCompute logview URLs. Disabled by default.

        If you enable this option, the logview URL is hidden in execution, data preview, and O&M logs for MAXCOMPUTE_SQL and logical table tasks that use account and password global variables. The URL is replaced with: The logview url is invisible because the current SQL is using the global variable "{dp_glb_xxx}", which is of type account and password..

        Note

        Available only for the MaxCompute compute engine.

      • Commit

        Parameter

        Description

        Automatically parse dependencies when committing offline development objects

        When enabled, the system parses and updates upstream dependencies on each commit of offline objects (SQL compute tasks, logical table tasks). Prevents missing dependencies.

        Validate field data types on logical table commit

        When enabled, validates that field calculation return types match the declared data type on logical table commit. Mismatches block the commit to prevent implicit type conversion errors.

      • Decommission and deletion

        Allow deleting published objects in the development environment: When enabled, published objects (compute tasks, integration tasks, logical tables, atomic metrics, business qualifiers, derived metrics) can be deleted from the development environment.

        Important

        Deleted objects cannot be recovered. Deleting an object in development without publishing the deletion makes the production object unmodifiable.

      • Default dependency period and policy

        Modify the Default Dependency Period and Default Dependency Policy.

        • Default Dependency Period: Options: Current period (day), Previous period (day), Last 24 hours, or N days ago. For N days ago, the default N is 2 (required).

        • Default Dependency Policy: Options: First instance, Latest instance, All instances, or Last instance.

        The initial default dependency periods and policies are as follows.

        Current node scheduling period

        Upstream node scheduling period

        Upstream self-dependency

        Default dependency period

        Default dependency policy

        Daily/Weekly/Monthly

        Daily

        Yes/No

        Current period (day)

        Last instance

        Daily/Weekly/Monthly

        Hourly/Minutely

        No

        Current period (day)

        All instances

        Daily/Weekly/Monthly

        Hourly/Minutely

        Yes

        Current period (day)

        Last instance

        Monthly/Weekly/Daily/Hourly/Minutely

        Monthly/Weekly

        Yes

        Current period (day)

        Last instance

        Monthly/Weekly/Daily/Hourly/Minutely

        Monthly/Weekly

        No

        Current period (day)

        Last instance

        Hourly/Minutely

        Daily

        Yes/No

        Current period (day)

        Last instance

        Hourly/Minutely

        Hourly/Minutely

        Yes/No

        Current period (day)

        Last instance

      • Tag values

        Manage task tag values. Click Add Tag to create a tag. Maximum: 50 tags, each up to 64 characters.

        Click Delete to remove a tag from all associated tasks. Re-adding the tag restores it.

    2. After you complete the configuration, click OK.

      Click Restore Defaults to reset.

    Table management settings

    Note

    Table management settings are not supported for StarRocks, GaussDB(DWS), Doris, or SelectDB.

    1. In the Table management settings area, click Edit to configure table deletion publish settings.

      • Automatically generate a pending publish item for table deletion after deleting a table using SQL: Enabled by default. Generates a pending publish item when a DROP TABLE statement runs in an ad hoc query or SQL compute task. When disabled, running DROP TABLE table_name does not generate a publish item.

      • Generate a pending publish item for table deletion when deleting a table in Table Management: Enabled by default. Generates a pending publish item when a table is deleted in Table Management. When disabled, no publish item is generated.

    2. Configure the Default Storage Format or Default External Table Storage Format. Supported formats vary by compute engine.

      Note

      • When the compute engine is AnalyticDB for PostgreSQL, Default Storage Format is not configurable.

      • Default External Table Storage Format is available only for MaxCompute.

      • In the table below, - indicates unsupported.

      image

      Engine default

      Hudi

      Delta (Delta Lake)

      Paimon

      Iceberg

      Kudu

      Parquet

      Avro

      Rcfile

      Orc

      Textfile

      Sequencefile

      Binaryfile

      Csv

      Text

      Json

      MaxCompute

      -

      -

      -

      -

      -

      -

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      -

      -

      -

      -

      Lindorm (compute engine)

      Supported

      -

      -

      -

      Supported

      -

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      -

      -

      -

      -

      Databricks

      Supported

      -

      Supported

      -

      -

      -

      Supported

      Supported

      -

      Supported

      -

      -

      Supported

      Supported

      Supported

      Supported

      Amazon EMR

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      -

      -

      -

      -

      • Transwarp TDH 6.x

      • Transwarp TDH 9.3.x

      Supported

      -

      -

      Supported

      Supported

      -

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      -

      -

      -

      -

      • CDH 5.x

      • CDH 6.x

      • E-MapReduce 3.x

      • E-MapReduce 5.x

      • Cloudera Data Platform 7.x

      • Huawei FusionInsight 8.x

      • AsiaInfo DP 5.3

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      Supported

      -

      -

      -

      -

    3. Configure the default lifecycle for physical and logical tables in MaxCompute. Default: empty (no lifecycle). Range: 1–36,500 days. Quick options: 7, 14, 30, or 360 days.

      Note

      Default lifecycle is available only for the MaxCompute compute engine.

    4. After you complete the configuration, click OK.

      Click Restore Defaults to reset.

    Default compute engine for standard modeling

    Set the default compute engine for standard modeling when using a Hadoop compute engine. Options: Hive, Impala, and Spark, with the following limitations:

    Important

    If a task type is not enabled for the project's compute source, the system defaults to Hive. Create a Hadoop compute source.

    • Hive: Cannot read Kudu source tables.

    • Impala: Reads Kudu source tables but does not support storing logical tables in Kudu format. Avoid Impala if your source tables are not in Kudu format.

      Note

      Impala is not available for Amazon EMR.

    • Spark: Cannot read Kudu source tables.

    Query acceleration

    Enable or disable MaxCompute Query Acceleration (MCQA). When enabled, MCQA applies to all ad hoc MaxCompute SQL queries and SQL unit queries in the analysis platform.

    Important

    Query acceleration is available only for MaxCompute.

    Single-tenant multi-engine

    In multi-engine mode, only Edit Lock, Storage Size Update Settings, Node Task Settings, and Table Management Settings are configurable. The following must be configured elsewhere:

    • Hide logview URL if SQL contains account and password global variables: Configure in Cluster Management > MaxCompute Cluster - Other Configurations. For more information, see Other configurations.

    • Default Storage Format/Default External Table Storage Format: Configure in Cluster Management > Other configurations. For more information, see Other configurations.

    • Default Compute Engine for Standard Modeling: Configure in Cluster Management > Other configurations. For more information, see Other configurations.

    • Query Acceleration: Configure in Cluster Management > MaxCompute Cluster - Other Configurations. For more information, see Other configurations.