BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Microsoft continues to flesh out its big data analytics tools to lure more traditional customers to its public cloud.
Microsoft recently made its SQL Data Warehouse service generally available, ending more than a year of beta testing, though it arrived several years after Amazon and Google began to offer similar capabilities with Redshift and BigQuery, respectively. Despite those delays, the Azure big data service, in many ways, leapfrogs what frontrunner Amazon offers, filling in a final piece of Microsoft's analytics platform and ecosystem to help IT shops shift data management to the cloud.
Azure SQL Data Warehouse represents the second generation of fully managed, cloud-based data warehousing, alongside services such as BigQuery, Snowflake and Panoply that separate the consumption of storage and compute resources, said Adam Ronthal, research director at Gartner. Amazon Redshift and IBM dashDB still sell well, but they represent more of a first-generation model that needs some refactoring to keep pace with newer options.
"In many ways, AWS [Amazon Web Services] Redshift falls short of the true vision for as-a-service," Ronthal said. "It still takes tuning and skills to work properly."
SQL Data Warehouse just became generally available, but it's a relatively mature platform, based off a parallel on-premises service acquired in 2008 with DATAllegro. While the beta lasted longer than usual, Microsoft was right to take its time to rearchitect it for the cloud, Ronthal said.
Azure big data tools an on-ramp from on premises
Milliman, an actuarial services provider based in Seattle, built a platform-as-a-service offering on Azure for clients to create software-as-a-service applications. Its customers can scale thousands of cores at a time, and it uses a host of analytics tools to manage its workloads, including HDInsight, Azure SQL, Data Factory, Azure Batch and Power BI Embedded.
The company works with SQL Data Warehouse and sees it as a natural extension of its ongoing data management efforts. It keeps tabs on Microsoft's competition, but with its deep commitment to Microsoft, it has no reason to jump ship yet, said Paul Maher, CTO of life technology solutions and principal at Milliman.
Paul MaherCTO of life technology solutions and principal, Milliman
"We're headed in the right direction to continue with Microsoft and the rate of innovation that's happening there," Maher said. "Being able to work closely with the engineering teams far outweighs anything we'd get from Amazon and Google."
Bringing business intelligence to the cloud was the biggest missing piece in Azure -- the lack of which made doing things such as data warehouse management in the cloud more difficult than it would be on premises, Maher said. Milliman can now pull together the various services by using Power BI with on-demand analytics and reporting, and it also opens the door to newer capabilities, such as SQL Data Warehouse and machine learning.
Ridge Tool Co., a hand tool manufacturer, uses HDInsight and blob storage, but it has also used Azure to build some of its own tools, including machine learning scripts and Elasticsearch cluster. Ridge has some storage in AWS and looked briefly at Amazon's machine learning tools, but found they weren't as easy to configure, said Joey Kushinski, data analytics manager at the Elyria, Ohio, company.
Azure provides a standard feature set and plenty of flexibility to build your own tools, while the ability to scale up gives customers a low entry price to play around with its big data tools, Kushinski said. Ridge expanded its use of the platform as it got more comfortable. Having the Azure engineering team available for added support has been crucial to that transition.
"Sometimes, it feels like Amazon is more self-service, but in this space, it's hard to be self-service because there's not a standard yet and you still have that volatility," Kushinski said. "I could see where it could go that direction in the future, but it's nice having someone say, 'This is the best way to make this work in your environment.'"
Azure covers its bases in competition with AWS
Microsoft is building a relatively complete ecosystem in Azure, in much the same way that Amazon is with AWS, and other competitors, including Google and Oracle, are trying to do, Ronthal said. It's a shift from products intended to do a lot of different things to more pointed services that do specific tasks very well.
"Microsoft has done a very good job of building out that ecosystem," Ronthal said. "It's still maturing, but if you look at the cloud market today, it's essentially a two-horse race with Amazon or Azure."
Microsoft has to expand its scope and reach, continue to push openness and get away from the image that it caters primarily to Windows environments, Ronthal said. Still, that existing customer base represents a big opportunity.
TierPoint, an IT services provider and Microsoft reseller in St. Louis, sees enterprise clients being two to three times more likely to be interested in Azure than AWS, said Shea Long, senior vice president of products. Microsoft is tapping into its enterprise licensing and comfort with CIOs to extend those relationships to Azure, and the SQL services play a big role in that.
"Microsoft wants to drive SQL as a focal point," Long said. "While people are embracing Hadoop, NoSQL and other database architectures, I think they want Microsoft SQL to be that front door."
Ziosk, a Dallas-based tabletop ordering and payment service for restaurants, has also expanded its use of Azure based on its analytics needs. The company has a sizable investment in an on-premises SQL database, but as that technology ages, it's looking carefully at migrating fully to Azure to take advantage of things like machine learning, SQL Data Warehouse, Data Lake and other internet of things services, while avoiding some of the transition and egress issues it has currently.
"With the investment Microsoft is making, it's becoming more and more interesting as an alternative," said Kevin Mowry, director of engineering at Ziosk.
Trevor Jones is a news writer with TechTarget's data center and virtualization media group. Contact him at email@example.com.
Top Azure management tips so far in 2016
Terms to know when using Google big data services
Microsoft delays Azure Stack release