Microsoft Releases SQL Server 2025 RC0 with TLS 1.3, AI Upgrades, and Ubuntu 24.04 Support

Microsoft publishes SQL Server 2025 Release candidate with TLS 1.3, Ubuntu 24.04 support and many AI improvements. Reading processes are up to 50 times faster, but an important restriction remains with Ubuntu support.

SQL Server 2025 RC0 is ready

Microsoft published the RC0 – the first release candidate – for SQL Server 2025. The new version ties in with the public preview of May 15, 2025 and brings a number of innovations. Microsoft has that in the TechCommunity announced. In addition to expanded development functions, performance, security and analysis skills are said to have been noticeably improved.

RC0 introduces a number of new functions, including the official support for Ubuntu 24.04 in development and test scenarios and the standard activation of TLS 1.3. The support for Ubuntu 24.04 is currently limited to the Enterprise Evaluation Edition, which is available for 180 days.

Preview features and AI improvements

According to Microsoft, the company tightens security with TLS 1.3 as the new standard. This version of the protocol offers greater encryption and more efficient hand shake and ensures better protection of data traffic between databases and applications. For production use, Microsoft is currently still recommending SQL Server 2022 in combination with Ubuntu 22.04 or RHEL 9. The highlight of this version is a new setting for developers and database administrators who want to try out experimental functions via an integrated Preview_FeaTures database configuration.

According to Microsoft, this opt-in mechanism enables new functions to explore in a controlled and flexible manner. The AI ​​functions were significantly expanded. Developers can create vector indices via the Preview_feaTurses function and carry out vector searches. The new version contains integrated vector bedding support with optimizations in both the .NET and the JDBC drivers.

Version mainstream support end extended support end

The improvements should lead to significant increases in performance: reading processes are up to 50 times faster, writing processes more than three times faster and Bulk copy processes would be carried out 19 times faster.

Such improvements are particularly relevant for workloads with machine learning, search applications and generative AI, where speed and scalability are decisive. How and when it continues is not yet known, since Microsoft has not published a official schedule for the release for SQL Server 2025.

Amit Gohar