Microsoft's Azure platform looks like, at first glance, a bewildering array of services and features, many of which have been recently updated including Azure Stream Analytics, SQL Database, Container Instances, SQL Data Warehouse, and others. While these are useful on their own, when interconnected, they become even more useful. That's what has just happened with Azure Data Lake Store (ADLS), SQL Server Data Tools (SSDT), and Azure Analysis Services (AAS).
For those not familiar, Analysis Services is an Online Analytics Processing engine - basically the online equivalent of SQL Server Analysis Services -, SSDT is a series of tools which allow you to "build SQL Server relational databases, Azure SQL databases, Integration Services packages, Analysis Services data models, and Reporting Services reports", and Data Lake Store is a scalable storage and analytics offering from Microsoft.
Today's announcement brings all three together, in the sense that you can now improve big data analytics workflows via "rich interactive analysis" with Analysis Services "for selected data subsets" in Azure Data Lake. The image below offers an overview of how the services interconnect and what the different outputs are:
As can be seen, this trio of services lets you "consume Azure Analysis Services models in Microsoft Power BI, Microsoft Office Excel, and Microsoft SQL Server Reporting Services". In addition to this, Data Lake's own Analytics component can run U-SQL batch jobs against source data, thus generating less overhead when importing the files into AAS.
Microsoft also points out that while the integration of AAS, SSDT and ADLS exists, it's very quick to process whatever job you throw at it:
Exporting approximately 2.8 billion rows of TPC-DS store sales data (~500 GB) into a CSV file took less than 7 minutes and importing the full 1 TB set of source data into Azure Analysis Services by using the Azure Data Lake connector took less than 6 hours.
An exact rundown of how to take advantage of the new capabilities announced can be found on the Analysis Services Team Blog.
