Microsoft hasn’t been shy about its goals for the data-driven age. It plans to bring business intelligence to a billion screens and to remove the barriers that are preventing broad adoption of advanced analytics.

These are lofty ambitions, perhaps, but there is no company on the planet that’s in a better position to deliver on them.

Put aside, for a moment, that Excel is a default tool for data crowd, that Office 365 is a next logical step for workers, and that so many enterprises inherently trust the Azure Cloud. The company is also rolling out big data solutions that are, at once, as powerful and compelling as those that scrappy start-ups are delivering today.

Latest Updates

In just the past year it's introduced HDInsight on Azure, born-in-the cloud SQL 2014 with in-memory capabilities, Big Data in a Box (which allows you to run queries across both traditional relational data warehouses and relational data stored in Hortonworks’ impressive Hadoop distribution), AzureML (Microsoft’s Machine Learning Service) and Power BI in the Cloud.

Today Microsoft announces a series of new capabilities that make it easier for Enterprises to run their businesses smarter by leveraging more of their data assets.

First on the list is an update to Microsoft Azure HDInsight – its cloud-based distribution of Hadoop. Customers can now process millions of Hadoop events in near real time. Eron Kelly, general manager of the company’s data platform division, said that the company is now previewing support for Apache Storm clusters in Azure HDInsight.

What this means for knowledge workers is that they’ll be able to gain insights from places like the Internet of Things (IOT) as events occur. Take for example an elevator in a 50 story building that begins to at an outlier speed. Knowing that this is happening, in real time, may provide an opportunity to address the problem before 30 workers get stuck on it.

New Features, Support

Microsoft will leverage Hortonworks’ HDP 2.2 (available next month) which includes support for hybrid Hadoop data clusters between on premise and the Azure cloud. Customers who choose to leverage its capabilities will have low cost option to back up their data on the Cloud. Not only that, but they’ll also be able to scale, from 100 nodes on premise to 1000 nodes on Azure when the need arises.

Machine Learning for the Azure Marketplace

Microsoft has added sample Machine Learning solutions to the Azure marketplace which can speed time to delivery on applications like anomaly detection (such as in the elevator example), figuring out what items and services retailers might suggest to customers by providing “frequently bought together” data, and a recommendation engine that Microsoft uses in its own Xbox business.

Kelly said that the gates are wide open for data scientist s to bring their own solutions to the marketplace.

At the end of the day, Microsoft is working to make Azure the best cloud for not only data storage, but a home for processing data, working with predictive analytics, and ultimately helping customers make better decisions so that they can serve their own customers better as well.