Skip to main content
Documentation
Documentation
Adaptive Planning
Last Updated: 2026-10-16
Troubleshooting: Slow Data Loader With Subquery Columns

Troubleshooting: Slow Data Loader With Subquery Columns

Subquery columns often generate correlated SQL that runs repeatedly for each row in the staging table. Runtime increases when:
  • More rows match the subquery condition.
  • Related tables grow in size.
  • Subqueries depend on SQL columns that can't be indexed.
  • Downstream join tables build on a slow base table.
Select the
Bypass Data Import
 check box on the slow data loader.
  1. Identify the slow loader in the task log.
    1. Select
      Integrations
       from the main menu.
    2. Select
      Design Integrations
      .
    3. Select your task in the
      Component Library
      .
    4. Select
      Logs
       in the
      Actions
       panel.
    5. Identify the slow data loader based on the logs.
  2. Select the
    Bypass Data Import
     check box.
    1. Select the slow data loader in the
      Component Library
      .
    2. Select
      Preview loader output
       in the
      Actions
       panel.
    3. Select the
      Bypass Data Import
       check box.
    4. Select
      Preview loader output
      .
Identify and refactor the slow staging table.
If the issue remains after you use the
Bypass Data Import
 check box, open the staging table and time the Preview Data process to find the slow base table to refactor.
  1. Select the target data source for the slow data loader in the
    Component Library
    .
  2. Select the target staging table.
    If the target table is a join or union table, trace back to the first slow base table.
  3. Select
    Preview data
    .
  4. Repeat these steps until you identify the slow table.
  5. After you identify the slow table, check the table for:
    • Subquery columns.
    • SQL columns used in joins.
    • High data volume.
    • Duplicate rows from a cartesian join.
  6. Refactor your staging table to:
    • Use a join-based design.
    • Filter earlier.
    • Add indexes on base columns.
    • Prepare data upstream.