In a significant milestone for the 2024 Stack Overflow developer’s favorite relational database management system (DBMS), the PostgreSQL Global Development Group has released PostgreSQL 17. This latest version of the open-source DBMS introduces a host of new features and performance improvements that further solidify PostgreSQL’s position as a leading database solution for enterprises and developers alike.
Also: 20 years later, real-time Linux makes it to the kernel – really
One of the most anticipated new features in PostgreSQL 17 is native support for incremental backups. Previously, you had to use third-party programs for this; now, it’s baked into the server. This should reduce storage requirements and recovery times, making it particularly valuable for enterprise database environments where full backups are time-consuming and resource-intensive.
PostgreSQL 17 significantly expands its SQL/JSON capabilities, introducing new functions like JSON_TABLE(), JSON_EXISTS(), JSON_QUERY(), and JSON_VALUE(). These additions bring PostgreSQL more in line with the SQL:2023 standard, offering developers powerful tools for interacting with JSON documents in a SQL-friendly manner.
<!–>
The new PostgreSQL also introduces the ability to configure the Simple Least Recently Used (SLRU) cache, which is crucial for handling subtransactions. This feature allows applications with high transaction volumes to run more efficiently without requiring significant rework of their transaction handling.
PostgreSQL also – finally – improved logical replication, which enables you to replicate data between PostgreSQL databases. Previously, if you used logical replication to start running a standby database in case of a failure, you had to resync the replication. This ate up a lot of time. Database administrators won’t have to deal with this delay from this release on.
“Our work on PostgreSQL 17,” noted Jozef de Vries, chief product engineering officer of EnterpriseDB, a leading PostgresSQL company, “underscores a belief that Postgres isn’t just a database – it’s a data platform supporting your most critical business systems. By introducing features like incremental backups and advancing features like JSON functionality and performance, EDB is doubling down on our commitment to the open source project while enabling enterprises to ‘just use Postgres’ to solve their most pressing data challenges.”
Also: Linux was never made for the cloud, but DBOS is – and you can try it for free
In addition to new features, PostgreSQL 17 comes with several performance enhancements. The most significant of these are:
- Improved handling of write-ahead log (WAL) locks, with some tests showing up to a 2x performance improvement for workloads with highly concurrent changes.
- Optimization of queries using the IN clause with B-tree indexes. These are PostgreSQL’s standard index type. This will provide significantly improved execution times.
- Support for Intel AVX-512 chip instructions for certain functions.
With its focus on advanced features and better performance, PostgreSQL 17 represents a substantial step forward for the database platform. As organizations rely increasingly on data-driven decision-making and AI applications, the enhancements in this release are likely to cement PostgreSQL’s position as a go-to DBMS for modern database needs.
