C3 AI Documentation Home

Rebuild Application Dependencies

After loading historical or backfilled data, downstream artifacts such as stored calculations, hierarchy tables, and cached metrics may become stale or incomplete. The Rebuild Application Dependencies feature allows data engineers to recompute these derived artifacts to ensure that the application reflects the latest data.

This feature provides a guided workflow in the Data Integration interface for triggering an application rebuild without using the developer console.

When to Rebuild Application Dependencies

You should rebuild application dependencies in the following scenarios:

  • After loading historical or backfilled data

  • After bulk ingestion jobs that populate large volumes of records

  • After schema or dependency changes that affect calculated fields or metrics

  • When dashboards, metrics, or derived objects appear out of date

Running a rebuild ensures that all derived artifacts are recomputed and consistent with the current data.

Open the Rebuild Application Dependencies Dialog

  1. Open the Data Integration canvas.

  2. Click the Settings icon.

  3. Select Rebuild Application Dependencies.

The Rebuild Application Dependencies dialog opens.

Configure the Rebuild

You can configure the rebuild using a preset configuration or by manually selecting dependency types.

Configuration Presets

The dialog provides several presets to simplify common rebuild scenarios.

PresetDescription
Post-Historical LoadRecommended after loading or backfilling historical data. Rebuilds common derived artifacts such as calculated fields, hierarchies, timeseries normalization, metric dependencies, and cached metrics required for consistent application behavior.
Complete RebuildRebuilds all supported dependency types including stored calculations, hierarchies, timeseries normalization, metrics, analytics, callbacks, ACLs, and compaction. Useful for a full refresh of application artifacts.
Calculations OnlyRecomputes stored calculations without rebuilding other artifacts.
CustomAllows manual selection of dependency types to rebuild.

If you manually modify dependency selections, the preset automatically switches to Custom.

Dependency Types

Each dependency type corresponds to a class of derived artifacts that may need to be refreshed after data ingestion.

Dependency TypeDescription
Stored CalculationsRecomputes persisted calculated fields that depend on loaded data.
Hierarchy DenormalizationRebuilds denormalized hierarchy tables used for rollup queries.
Timeseries NormalizationConverts raw timeseries data into optimized query structures.
Metric DependenciesRebuilds the dependency table that tracks which source objects invalidate metrics.
Cached MetricsRefreshes cached metric values used by dashboards and reports.

Select the dependencies that should be rebuilt.

[!NOTE] The available dependency types may vary depending on configuration and are displayed in the UI.

Configure Batch Sizes (Optional)

Advanced users can configure batch sizes for each dependency type to optimize rebuild performance based on data volume and available system resources. Batch sizes determine how many records are processed in each operation during the rebuild.

To modify batch size settings:

  1. Open the Rebuild Application Dependencies dialog.

  2. Select a preset or choose Custom.

  3. Expand the Advanced section.

  4. Update Batch Size and Sub-Batch Size values as needed.

  5. Start the rebuild.

If no changes are made, the system uses default values.

Monitor Performance

After starting a rebuild:

  1. Navigate to the Batch Jobs page.

  2. Locate the rebuild job.

  3. Review execution time, logs, and resource usage.

  4. Adjust batch sizes if needed and rerun the rebuild.

Start the Rebuild

  1. Choose a configuration preset or manually select dependency types.

  2. Click Start Rebuild.

The system starts a background rebuild job that recomputes the selected dependencies.

Only one rebuild job can run at a time. If a rebuild is already running, the dialog displays a notification and provides options to view the job status or cancel the running rebuild.

Monitor Rebuild Progress

Rebuild operations run as background Batch Jobs.

To monitor the progress of a rebuild:

  1. Navigate to the Batch Jobs page.

  2. Locate the rebuild job.

  3. View job status, logs, and completion details.

Cancel a Running Rebuild

If a rebuild is already running, the dialog displays a notification indicating that an application rebuild job is in progress.

You can:

  • View Current Status to open the job in the Batch Jobs page

  • Cancel Running Job to stop the rebuild

After the job is canceled or completed, you can start a new rebuild.

Best Practices

  • Run a rebuild after large historical data loads to ensure derived artifacts remain consistent.

  • Use Post-Historical Load for typical backfill scenarios.

  • Use Complete Rebuild only when a full refresh of application artifacts is required.

  • Monitor rebuild jobs through the Batch Jobs page to verify completion and identify potential issues.

Was this page helpful?