Which of the following is an example of a case where you would NOT want to use an Automation in FME Flow?

• A. Automatically processing data that has been sent in through email
• B. Monitoring a directory for newly uploaded files
• C. Alerting an administrator when a database update fails
• D. Handling data streamed from a temperature sensor more than once per second

Answer: (D)

Automations in FME Flow are designed for event-driven, discrete tasks that respond to a state change or specific trigger. They work well when something happens sporadically, like a new file arriving, an email arriving, or a condition being met in a database. A temperature sensor reporting data more than once per second is a high-frequency, continuous stream. Triggering an automation for every sample would generate a flood of runs, create overhead, and likely fail to meet real-time processing needs. For that kind of streaming data, you’d use a continuous or streaming ingestion path rather than discrete automations. The other scenarios fit automation well because they involve identifiable, less frequent events that can reliably trigger a workflow: processing data from an email, detecting new files in a directory, or alerting an administrator when a database update fails.