At the Microsoft Ignite conference this week, Microsoft gave attendees a glimpse of the next version of the Microsoft SQL Server 2016 database.
Tiffany Wissner, senior director, data platform marketing at Microsoft, said that while Microsoft has not committed to a ship date yet for Microsoft SQL Server 2016, it wants IT organizations to start thinking now about how the convergence of transaction processing and analytics running simultaneously in memory will influence their enterprise application strategies.
To accomplish that goal, SQL Server 2016 now includes a columnstore that runs on top of the row-based transaction processing engine. Providing that capability should result in transaction processing and analytics applications that run 100 times faster than they do today on Microsoft SQL Server 2014.
Other major enhancements include a new Stretch Database feature that enables rows to be extended across a hybrid implementation of the database. Wissner says IT organizations can deploy a portion of the Microsoft SQL Server on premise alongside another instance running on Microsoft Azure that to an applications looks like one unified database. Wissner says this implementation will be especially useful to applications that increasingly make use of a large number of database tables.
Microsoft is also adding a new Always Encrypted capability that can be applied to data both at rest and in motion. Because the data is always encrypted, Wissner notes that all the overhead associated with encrypting and decrypting data is no longer generated. Other security enhancements include support for row-level security and dynamic data masking.
Finally, Microsoft also revealed that SQL Server 2016 will support the Javascript Object Notation (JSON) data format and the open source R programming language.
While there is a lot of fierce debate these days over in-memory database architecture, Wissner says that Microsoft is committed to making SQL Server 2016 as backwards-compatible as possible with existing applications versus requiring developers to rewrite their applications entirely.
Regardless of the way IT organizations go forward, transaction processing and analytics within the same application are about to joined together in memory now and forever.
