Incremental Model Merge Performance Degradation

warning

latencyUpdated Mar 2, 2026

Identifies incremental models experiencing slow merge operations due to large data volumes, inadequate clustering, or missing optimization features.

Sources

How to Debug dbt Model Failuresoneuptime.com

Optimize and troubleshoot dbt models on Databricks - dbt Docsdocs.getdbt.com

Technologies:

dbtSymptoms of this issue are visible in dbt metrics and logs

dbt.model.incremental.merge_duration

dbt.model.execution_time

dbt.model.rows_affected

How to detect:

Track execution time trends for incremental models, comparing merge duration to historical baselines. Alert when merge operations exceed 2x the historical average or when incremental builds take longer than full-refresh equivalents.

Recommended action:

Analyze the incremental logic in the model's is_incremental() block. Implement Z-ordering on high-cardinality columns, enable auto-compaction, run OPTIMIZE and ANALYZE TABLE commands, or adjust the incremental window size. Review compiled SQL for merge efficiency.