← All updates Databricks Update One-Pager Light version
databricks Platform · Updated Oct 14, 2025

Create backfill job runs

Job backfills allow you to trigger job runs to backfill data from the past.

Update Open docs ↗release note ↗
What it does

Job backfills allow you to trigger job runs to backfill data from the past. This is useful for loading older data, or repairing data when there are failures in processing. After you have created a job that runs on a schedule, there are times when you need to backfill data into the system.

Why it matters

By using the same time interval your regular job, the performance and optimizations that you have set up in your job are maintained.

Limitations · remember
!Pipeline tasks are not parameterized. The pipeline is run as-is.
!Pipeline tasks can't be run concurrently, so jobs with pipeline tasks are run sequentially rather than in parallel. A warning is shown in the Run backfill dialog when this happens, and in Settings, if you set the maximum concurrent runs greater than 1 when there is a pipeline task in the job.
Architecture
Databricks Platform → Platform
In short

Job backfills allow you to trigger job runs to backfill data from the past. This is useful for loading older data, or repairing data when there are failures in processing.

Read the full documentation ↗
Key takeaway
Job backfills allow you to trigger job runs to backfill data from the past.
2025-10-14-create-backfill-job-runs Generated by the Databricks Updates Agent · browse all updates