Introducing Azure Stream Analytics


We are so excited today to be announcing to you Azure Stream Analytics – a new fully managed stream processing service in the cloud. Stream Analytics provides low latency, real-time processing of millions of events per second in a highly resilient service. This morning at TechEd Europe 2014, Microsoft officially announced the preview of Azure Stream Analytics.

Stream Analytics enables you to easily combine streams of data with historic records or reference data to derive business insights quickly and easily, by providing a range of operators from simple filters to complex correlations. Defining time based windowed operations such as windowed aggregates, or correlating multiple streams to detect patterns such as sequences, or even comparing current conditions to historical values and models, can be done in a matter of minutes using the simple set of SQL-like Stream Analytics Query Language operators. Specifying a streaming pipeline is as simple as configuring its inputs and outputs, and providing a SQL-like query describing the desired transformations. While this suffices for most simple cases, higher-scale and more complex scenarios will also benefit from Stream Analytics configurability and the ability to determine how much compute power to dedicate to each step of the pipeline to achieve the desired peak throughput.

The Stream Analytics service is optimized to provide users very low cost to get going and maintain real time analytics solutions. The service is built for you to pay as you go based on usage. The usage is derived based on the volume of events processed and the amount of compute power provisioned within the cluster to handle the respective streaming jobs.

To get started, check out the Stream Analytics service page and visit the Microsoft Azure Portal.

– The Azure Stream Analytics Team

Comments (2)

  1. Roberto Messora says:

    I'm testing Stream Analytics and I noticed that the JSON data must be flat: I mean, a JSON data with only one level of properties. I'm trying to analyze a event hub stream that manages complex JSON messages, but it seems that the Query Language doesn't support properties dot notation.

    Is it correct?

  2. Hi Roberto, you're correct that nested data isn't yet supported.  It's on our roadmap and you can vote for the feature on the Azure Feedback Forum: feedback.azure.com/…/6704733-provide-support-for-complex-nested-data-types-in

    Thanks for the feedback!

Skip to main content