Stream Analytics Key Concepts

Flows​

In Stream Analytics, a computational task is represented as a flow, which outlines the different elements involved in the process. The flow encompasses the streaming data sources as the input Node, the computing logic implemented through Compute Nodes, and the sinks or actions as the output Node.

Once the flow is defined and deployed, it operates continuously. It continuously fetches data from the specified sources, performs calculations based on the defined Compute logic, and then triggers the specified actions, utilizing the obtained results.

Sources​

Sources are employed to retrieve data from external systems, and they come in two types: unbounded streaming data referred to as "stream" and lookup data. When integrating a source into a flow, it is necessary to include at least one stream as part of the configuration.

The source's primary function is to establish a connection with an external resource and fetch data from it in a streaming manner. Once the data is obtained, common tasks like decoding and transforming it according to the schema can be accomplished by configuring the appropriate properties.

Sinks​

Sinks serve the purpose of writing data to an external system. They can be utilized for writing control data to trigger specific actions, as well as for saving status data to external storage.

Within a flow, sink types are employed as actions. Rules have the flexibility to contain multiple actions, and these actions can involve the same sink type.

Jobs​

Jobs in Stream Analytics is essentially a pre-arranged flow scheduled to be executed at specific times. They facilitate automated and efficient task management within various computing systems.

Point of Interest​

A Point of Interest (PoI) refers to a geographical location represented as a point, circle, or polygon with coordinates. It serves as input for Stream Analytics Geospatial functions.