Hands-On Data Warehousing with Azure Data Factory
上QQ阅读APP看书,第一时间看更新

Pipelines

Pipelines have been modified quite a lot in V2. They don't have any windows of execution, with start times and end times. Pipelines can now be executed using the following technique:

  • On demand via .NET, PowerShell, REST API, or Python
  • Trigger:
    • Schedule trigger: This trigger uses a wall clock kind of schedule, for example, a pipeline can be executed on a weekly basis every Tuesday and Thursday at 10:00 AM
    • Tumbling window trigger: This works on a periodic interval, for example, every 15 minutes between two specific dates