Today at TechEd Europe 2014, Microsoft announced the preview of Azure Stream Analytics. Stream Analytics is a real-time event processing engine that helps uncover insights from devices, sensors, infrastructure, applications, and data. With out-of-the-box integration to Event Hubs, the combined solution can both ingest millions of events as well as do analytics to better understand patterns, power a dashboard, detect anomalies, and kick off an action while data is being streamed in real-time. Powered by the Azure cloud, Stream Analytics is highly scalable and has no hardware or other up-front cost, installation or set-up to do. Developers simply describe their desired analytic function in a SQL-based query syntax and the system will transparently deal with distribution for scale, performance, and resiliency.
Building from the announcement we made at Strata+Hadoop World around Apache Storm preview availability in Azure HDInsight, Stream Analytics is also a stream processing engine. By making both Storm and Stream Analytics available, Microsoft is giving customers options to deploy their real-time event processing engine of choice.
Azure Stream Analytics brings a unique perspective in the overall complex event processing and real-time processing space with the following customer benefits:
- Low cost: Stream Analytics is architected for multi-tenancy meaning you only pay for what you use and not for idle resources.
- Faster developer productivity: Stream Analytics allow developers to use a SQL-like syntax that can speed up development time from thousands of lines of code down to a few lines. The system will abstract the complexities of the parallelization, distributed computing, and error handling away from the developers.
- Elasticity of the cloud: Stream Analytics is built as a managed service in Azure. This means customers can spin up or down any number of resources on demand. Customers will not have to setup costly hardware or install and maintain software.
For more information on Azure Stream Analytics: