Delta Live Tables, now called Lakeflow Declarative Pipelines, charges for the compute used to run pipeline updates. The DBU rate depends on the pipeline mode and plan.
Pipeline cost depends on update frequency, source volume, transformations, cluster size, and the number of tables refreshed. A pipeline that runs more often can increase cost without changing its SQL.
Measure DBUs, cloud infrastructure, runtime, and data volume for each update. Separate development and production pipelines, and use the pipeline settings that match the required latency. Databricks publishes the applicable rates on its pricing page.