The SQL Server 2019 database platform includes a broad range of technologies, features, and services, supporting mission-critical applications such as analytics, in-memory databases, business intelligence (BI), and reliable and scalable online transaction processing (OLTP). The SQL Server platform has acquired capabilities to handle data integration, data warehousing, reporting, high-speed advanced analytics, data replication and programmability features including hosting in-database common language runtimes, service broker hosting, and semi-structured datatype management. Microsoft realizes that with such a breadth of services available in the platform, not all customers or applications need every feature enabled on every SQL Server instance. In many cases, it is preferable to separate SQL Server services either through virtualization or by managing service-specific implementation.
Key new features and updates in SQL Server 2019 include:
- SQL Server on Linux
- Supports participation in transactional, merge, and snapshot replication topologies in the publisher, distributor, or subscriber roles
- Is configurable for user authentication through Microsoft Active Directory and for replication and distributed queries
- Supports participation in Availability Groups that are authenticated by Active Directory, in addition to the previously supported certificate-based authentication
- Can initiate and participate in distributed transactions through the Linux version of the Microsoft Distributed Transaction Coordinator (MSDTC), which can also participate in distributed transactions, including developer support, with other third-party transaction coordinators
- Big Data clusters (new feature)
- Support Spark Linux containers on Kubernetes
- Integrate with data stored on HDFS
- Enable advanced analytics and machine learning on Spark
- Use Spark streaming to ingest data into persistent SQL Server storage—SQL data pools
- Use query books that provide a notebook experience in Microsoft Azure Data Studio
- Use Sparklyr R interface
- Use Jupyter notebooks to assist with deployment and discovery, diagnosis, and troubleshooting for components in a SQL Server Big Data cluster
- Always-On Availability Groups
- Support up to five synchronous replica pairs—one primary and up to four secondary replicas—with automatic failover between replicas
- Enable high-availability configurations for SQL Server running in containers, using Kubernetes as an orchestration layer
- Azure Data Studio
- Provides a lightweight, open source, cross-platform desktop tool for the most common tasks in data development and administration
- Connects to SQL Server on premises and in the cloud from Windows, macOS, and Linux
- Polybase
- Supports using external table column names for querying SQL Server, Oracle, Teradata, MongoDB, and ODBC data sources
- Machine Learning Services
- Enables running R and Python scripts on Linux
- Can be installed on a Windows Server failover cluster for high availability
- Uses existing R or Python scripts without modification to process data at the table partition level to train a model for each table partition and parallelize model training per partition
- Power BI Report Server
- Enables scaling to thousands of users
- Enables reports that are designed in Power BI Desktop to be deployed on an on-premises server rather than from the Power BI cloud service
For more details about Microsoft SQL Server 2019, see the SQL Server 2019 CTP announcement archive and the Microsoft SQL Server 2019 technical white paper.