Microsoft Extends Big Data Platform On Azure
At the Microsoft Build 2015 conference today, Microsoft (MSFT) unfurled a range of elastic database-as-a-service offerings designed to make multiple database implementations be addressable as a single logical entity.
Renga Rengarajan, corporate vice president for the Microsoft Data Platform, says the Azure SQL Database elastic database simplifies running applications against hundreds of instances of Microsoft SQL databases running in the Azure cloud.
In addition, Microsoft also announced Azure SQL Data Warehouse, which makes use of the massively parallel database functionality that Microsoft developed for SQL Server to create a data warehouse in the cloud that can dynamically scale up and down as needed.
Finally, Microsoft also launched Azure Data Lake, a distributed implementation of a file system that is compatible with the Hadoop Distributed File System (HDFS).
While the elastic implementations of Azure SQL Database are available today, the Azure SQL Data Warehouse is only accessible via a public preview program. Azure Data Lake, meanwhile, is only available as part of a private preview program. Pricing for the various database services, said Rengarajan, will be calculated using an average workload formula with each individual customer.
Collectively, Rengarajan said Microsoft is putting in place a framework that will enable any application to access Big Data resources in the cloud. Rather than limiting the size and scope of their applications, Rengarajan said developers can now think in terms of building applications where data streams into applications in real time to continuously update the end user experience.
On top of that framework Microsoft also envisions developers composing applications that make use of machine learning software that Microsoft plans to expose via a common set of application programming interfaces (APIs) running on the Microsoft Azure platform.
In fact, Rengarajan said the biggest challenge Microsoft now faces in the cloud is the limitations that developers impose on themselves because they have internalized how legacy data management platform have historically functioned. Microsoft Azure, said Rengarajan, is essentially an intelligent cloud where all the boundaries between various data sets have been blown away.
Rengarajan adds that Microsoft fundamentally assumes that developers will be invoking those data sets across hybrid cloud computing environments now and forever. As such, solution providers across the channel should expect to be support federated data management platforms residing both inside and out of the Microsoft Azure cloud.
The degree to which Microsoft can leverage its database expertise and huge base of developers to claim the hybrid cloud computing mantle remains to be seen. But given both data sovereignty and the natural forces of data gravity inside and out of the enterprise there’s no doubt that the Big Data management challenges the IT organizations will be looking to address will be quite substantial in the years ahead.
