X
Business

Microsoft's SQL Server Next for Linux, Windows hit public preview

Microsoft's next version of SQL Server is available in public preview form for both Linux and Windows. Microsoft also is introducing SQL Server 2016 SP1.
Written by Mary Jo Foley, Senior Contributing Editor

Microsoft is releasing a public preview of the next version of its SQL Server database, which will work on Linux and Windows, both.

sqlserveronlinuxpublicpreview.jpg

Microsoft released a private preview of SQL Server for Linux in March 2016. At that time, Microsoft officials said SQL Server for Linux was a subset of SQL Server 2016 for Windows and would work on Ubuntu or as a Docker image. (Microsoft later added support for Red Hat Enterprise Linux to the private preview.)

In March, company officials also said that users who bought SQL Server would be able to use that license on either Windows Server or Linux, but didn't go so far as to say that the version of SQL Server for Linux would be the same one that would be considered SQL Server Next for Windows, too. Microsoft only recently -- as of June 1, 2016 -- made its most recent version of SQL Server for Windows, SQL Server 2016, generally available.

However, as of today, Microsoft is positioning SQL Server Next as being both a Linux and a Windows thing.

Microsoft officials said SQL Server Next on Linux and SQL Server Next on Windows are based on a single code base. "There are some features that have platform specific characteristics. For example, file paths in Linux are different than file paths in Windows," said a spokesperson. But they are not two completely different products.

At its Connect() 2016 developer event in New York City on November 16, Microsoft officials reiterated that the coming SQL Server on Linux release would include "all major features of the relational database engine." That includes in-memory online-transaction processing (OLTP), in-memory columnstores, Transparent Data Encryption, Always Encrypted, and Row-Level Security, they said.

The public preview of SQL Server on Linux will be available on Ubuntu, Red Hat enterprise Linux, and SUSE Linux Enterprise Server. The public preview will be available "soon" on Azure Virtual Machines and on Docker Hub, Microsoft officials said today. Microsoft officials are still saying that SQL Server will be generally available for both Linux and Windows by mid-2017.

The public preview of SQL Server on Linux will be available for download today from Microsoft's SQL Server on Linux site. The public preview of the next release of SQL Server for Windows (technically known as Community Technology Preview, or CTP, 1) will be available from the TechNet Evaluation Center.

Microsoft execs also announced SQL Server 2016 Service Pack 1 at Connect() 2016. SP1 is available today, November 16, 2016.

SQL Server SP1 is bringing some of the higher-end capabilities that to this point were only available in the Enterprise version of SQL Server 2016 down to the Standard Edition. Among those capabilities that will be common across both Standard and Enterprise are in-memory OLTP, real-time operational analytics, data encryption at rest and in motion, and PolyBase big-data connectivity.

Here's a chart from Microsoft showing how the Enterprise and Standard versions of SQL Server 2016 will stack up, as of SP1.

In other data-management news, Microsoft is adding new features, including new machine learning functions, to its R Services inside SQL Server. Today's also the day Microsoft's R Server for Azure HDInsight (aka Hadoop on Azure) is generally available. Microsoft also allows VSCode developers to connect to SQL Server, including SQL Server on Linux.

And in conjunction with Microsoft Connect(), the company is making Azure Data Lake Analytics and Azure Data Lake Store both generally available as of today. Azure Data Lake is Microsoft's hyperscale repository for big-data analytic workloads in the cloud. Azure Data Lake Analytics and Data Lake Store are both derived from Microsoft's own internal "Cosmos" big-data storage and analytics service. ("Kona" was the codename for the analytics engine and "Cabo" for the storage engine.)

Editorial standards