Azure Stream Analytics Processes Events in Real Time

Azure Stream Analytics, which became available as a preview in Oct. 2014, is a hosted service that processes large volumes of streaming event data with low latency to help customers derive insights, draw conclusions, and trigger actions in real time. The service can be used to process data from sources such as Web site clickstreams, transactional systems, and equipment sensors. Stream Analytics could simplify deployment of and reduce costs of complex event processing relative to on-premises alternatives by leveraging the advantages of a Platform as a Service offering. However, the preview can only accept data from and send data to other Azure services.

High Volume Event Streams Handled

High volumes of event data, logged on the order of megabytes per second and higher, is commonly archived for later analysis (such as by a Hadoop implementation like Azure HDInsight), but critical value can be derived from incoming data streams by processing them as they arrive. For example, ad targeting, fraud and identity theft detection, stock market predictions, equipment control, and automobile telemetry-based services often require real-time interpretation of events to be most effective.