An output in ArcGIS Analytics for IoT is a result or action to be taken as the final step in a real-time or big data analytic. Analytics can emit data to a variety of different destinations, including storing data to a feature layer, sending an email, writing to a cloud store, and pushing to a third party system for device actuation. A real-time or big data analytic must have at least one output, but can send data to multiple outputs at the same time.
- An analyst would like to create a new feature layer to store sensor observations following processing in a real-time analytic.
- A researcher would like to append results of big data analysis to an existing feature layer each morning following processing in a scheduled (near real-time) big data analytic.
- An email with a relevant message and information should be sent from a real-time analytic if a feature meets a certain criteria.
Outputs in real-time analytics
Real-time analytics perform processing on data being ingested via a feed, analyzing each individual message as it is received. After processing is performed in the analytic, each feature is sent to an output individually. It is important to consider this when configuring certain kinds of outputs.
For example, if configuring a Email output, a new email would be sent for each feature sent to that output. For high velocity feeds, this may not be the desired behavior. As a best practice, use the Email output only for incidents that are expected to occur occasionally.
The following output types are available in real-time analytics:
Outputs in big data analytics
Big data analytics perform batch analysis on stored data. When the analytic completes its processing, features are sent to each output.
The following output types are available in big data analytics: