“It introduces a new paradigm where all joins, aggregations and machine learning are performed securely within the database itself without moving the data out, thereby enabling analytics on real-time transactions with great speed and parallelism,” he explained in a June 1 announcement. “As a result, analytical applications can now be far simpler and need only query the database for analytic results.”
In terms of crunching big data, SQL Server 2016 offers organizations in-memory processing features and the option of running SQL Server R Services as an in-database capability. R is the open-source statistical computing programming language that has proved popular among data scientists.Microsoft acquired Revolution Analytics, R’s leading commercial backer, in early 2015.On modern hardware, SQL Server 2016 will also offer significant performance gains, said Sirosh. Collaborating with Intel, Microsoft was able to a load a TPC-H-derived complex schema at a rate of 1.6 terabytes per hour on a 100TB data warehouse with a single, four-processor server (Intel Xeon E7). A complex query involving the entire 100TB database took 5.3 seconds, he claimed.Even previous-generation systems should notice brisker performance. “Simply by upgrading to SQL 2016, many customer workloads experienced large performance gains. For example, the 3TB TPC-H benchmark on the same hardware demonstrated a forty-eight percent performance gain,” according to Sirosh.To help keep sensitive business information out of the wrong hands, SQL Server 2016’s Always Encrypted feature protects data both in memory and at rest. Additionally, the software’s low-overhead Transparent Data Encryption capability, borrowed from Azure SQL Database, locks down user data by encrypting the entirety of a database’s storage.SQL Server 2016 is available now in four editions: Express, Standard, Enterprise and the free Developer Edition, the latter of which allows users to test the features found in SQL Server Enterprise.