Knowledge Base

Job Types Reference

Job Types Reference

This page provides a reference of all job types available in the Narrative Data Collaboration Platform.


Available Job Types

Job TypeDescription
materialize-viewMaterialized view creation and refresh jobs triggered by dataset creation, updates, or scheduled refreshes
forecastLegacy forecast jobs that approximate row counts for query definitions
nql-forecastNQL-based forecast jobs that estimate rows matching NQL query definitions
costsCost calculation jobs triggered by data processing estimation requests
datasets_sampleDataset sampling jobs triggered when requesting preview samples of datasets
datasets_calculate_column_statsColumn statistics calculation jobs triggered by data profiling requests
datasets_delete_tableDataset table deletion jobs triggered when cleaning up data resources
datasets_suggest_mappingsMapping suggestion jobs triggered by automated data integration processes
datasets_deliver_dataData delivery jobs triggered when connectors deliver data to external endpoints
delete-snowflake-tableSnowflake table deletion jobs triggered when cleaning up Snowflake resources
upload-statsStatistics upload jobs triggered when synchronizing statistical data and metrics
model_training_runLLM model training jobs triggered when starting fine-tuned training processes
models_deliver_modelModel delivery jobs triggered when deploying trained models to production data planes

Job States

All job types progress through these states:

StateDescription
pendingJob has been queued but not yet started
runningJob is currently executing
completedJob has finished successfully
pending_cancellationJob is being cancelled
cancelledJob was cancelled before completion
failedJob encountered an error and could not complete

Job Lifecycle and States

Understanding how jobs progress through states and execution patterns.

< Back
Rosetta

Hi! I’m Rosetta, your big data assistant. Ask me anything! If you want to talk to one of our wonderful human team members, let me know! I can schedule a call for you.