Process events with Azure Stream Analytics


Azure Stream Analytics is a platform-as-a-service (PaaS) event processing engine. It enables the transformation and analysis of large volumes of streaming data arriving from Azure Event Hubs and IoT Hub and static data from Azure storage. Using Stream Analytics, you can write complex time-based queries and aggregations over the data generated by connected sensors, devices, or applications. Stream Analytics processes the data in real-time, enabling powerful insights to drive real-time decision-making. A typical event processing pipeline built on top of Stream Analytics consists of the following four components:

  • Event producer: Any application, system, or sensor that continuously produces event data of interest. Examples include sensors tracking the flow of water through a utility pipe and an application such as Twitter that generates tweets against a single hashtag.
  • Event ingestion system: Receives the data from an event producer and passes it to an analytics engine. Azure Event Hubs, Azure IoT Hub, or Azure Blob storage can serve as the ingestion system.
  • Stream analytics engine: The compute platform that processes, aggregates, and transforms incoming data streams. Azure Stream Analytics provides the Stream Analytics query language (SAQL), a subset of Transact-SQL tailored to perform computations over streaming data. The engine supports windowing functions that are fundamental to stream processing and are implemented by using the SAQL.
  • Event consumer: A destination of the output from the stream analytics engine. The output can be stored in a data storage platform, such as Azure Data Lake Storage Gen2, Azure Cosmos DB, Azure SQL Database, or Azure Blob storage. Or, you can consume the output in near-real-time using Power BI dashboards.

Diagram of the event processing components available when building a pipeline with Azure Stream Analytics.

Operational aspects

Stream Analytics guarantees exactly once event processing and at-least-once event delivery, so events are never lost. It has built-in recovery capabilities in case the delivery of an event fails. Also, Stream Analytics provides built-in checkpointing to maintain the state of your job and produces repeatable results.

Because Azure Stream Analytics is a PaaS service, it's fully managed and highly reliable. Its built-in integration with various sources and destinations and flexible programmability model enhance programmer productivity. The Stream Analytics engine enables in-memory compute, so it offers superior performance. All these factors contribute to the low total cost of ownership (TCO) of Azure Stream Analytics.