Cortana Analytics: Microsoft's cloud analytics prix fixe
Ever since the first version of Power BI went into preview about two years ago, Microsoft has been on a cloud data and analytics tear. Microsoft's cloud-based Hadoop offering, HDInsight, went into general availability (GA) back in October of 2013 and Power BI made its own GA debut a few months later. Azure Event Hubs, Machine Learning and DocumentDB began their previews last summer and GA'd in November '14, February '15 and April '15, respectively. Azure Stream Analytics began its preview in October of last year and GA'd this past April.
And the pace hasn't let up. Azure Data Factory began a preview last October and this spring and summer brought us previews of Azure SQL Data Warehouse, Azure Data Lake and Azure Data Catalog. Finally, Power BI 2.0 -- a major reboot of the service -- has GA'd today.
Now what do we do?
That's a lot of stuff. Nonetheless, it's a bunch of separate services that Microsoft brought to market in relatively quick succession, rather than a truly integrated offering with multiple capabilities. Such an array of services can get complicated, both in terms of billing and implementation.
For example, Power BI, though it no longer requires an Office 365 subscription, is nonetheless a standalone service, not under the Azure umbrella. And even within the Azure services, there are a variety of billing mechanics, and two different portals that come into play, depending which services are used.
Where's the synchrony?
Ten years ago, Microsoft worked hard to coordinate what became the 2005 versions of its SQL Server database and Visual Studio developer suite, leading to a series of delays that frustrated developers. Perhaps as a result, the company now believes that separate, parallel development is the way to go. Integration can come later, first with branding and SKUs (i.e. the products that are sold or licensed), then later with shared user interface and technology.
Enter Cortana Analytics Suite. A brand and subscription bundle of all of the aforementioned cloud analytics products, combined with the Cortana digital personal assistant technology, as well as "Perceptual Intelligence," which includes computer vision, facial and speech recognition as well as text analytics. (Some of the Perceptual Analytics functionality is powered by technology from "Project Oxford," a Microsoft Research effort.)
Hey, Analytics
Cortana Analytics was announced at last week's Microsoft Worldwide Partner Conference, and the news was covered by ZDNet's Mary Jo Foley. General availability is slated for "later this fall" and promises to deliver a single subscription for all of the Azure Big Data and analytics services. Pricing will be disclosed in the fall as well.
Microsoft also promises to bring integrated vertical industry solutions to Cortana Analytics customers. These are essentially use case templates/accelerators, for industries the will likely include manufacturing, healthcare and financial services. While they may not be full-fledged products per se, and definitely won't constitute true integration of the services, they will nonetheless serve as canonical examples of how to use the services together.
Certain of the services have point-to-point integration already in place. Azure Data Factory has connectivity to Azure Stream Analytics, and the latter has connectivity to Event Hubs. Power BI knows how to talk to Apache Spark running on HDInsight. Azure Data Lake emulates HDFS (the Hadoop Distributed File System), which has native connectivity from the Power Query component of Power BI. Azure SQL Data Warehouse features Microsoft's PolyBase technology, which integrates HDInsight and other Hadoop distributions.
Show me
A live webinar on Cortana Analytics held yesterday and hosted by Microsoft's Corporate Vice President for Machine Learning, Joseph Sirosh, showed how far Microsoft is, and is not, along the path to tying all these services together. In addition to a lot of marketing slides and talking points, the team demonstrated some integration of Cortana and Power BI. In the demo, questions about data in Power BI were spoken into the Windows 10 Cortana user interface.Power View visualizations then appeared in the small Cortana window in response.
Cortana's speech recognition somehow displayed the text of the questions before the gentleman doing the demo finished reading them out. The demos may have been rehearsed previously, with Cortana effectively "auto-completing" the questions, or it may be that they were a bit contrived, shall we say. Regardless, the data visualizations coming back were almost certainly not staged, as such plain-English query capabilities, dubbed Q&A, have been part of Power BI for quite some time now.
The way to a cloud customer's heart is through his data
So the pieces may still be separate. But they are coming together and, it should be noted, comparable services from Amazon Web Services and Google Cloud Platform are rather loosely-coupled themselves.
Microsoft is serious about cloud analytics. Using a cloud platform for analytics brings with it significant use of that cloud platform's storage facilities, which begets further incumbency and revenue for that cloud platform. It's pretty clear that Microsoft gets this, and sees a single analytics subscription product as facilitating the payment of dividends on its gargantuan cloud infrastructure investment.
Though not without risk, this is a good bet on Microsoft's part. It will be fun to watch how Amazon and Google respond.
Post updated at 12:15pm Eastern Time to correct information on Power BI 2.0 GA, which was incorrectly described as taking place next week.