What’s New in PostgreSQL 17? Incremental Backup, SQL/JSON Boosts, and SLRU Cache

Read on: Highly optimized incremental backup, expanded SQL/JSON support, and a configurable SLRU cache are three new features in PostgreSQL 17.

The PostgreSQL global development team officially released Postgres 17, marking another milestone for the community and its contributors. As the top‑ranked DB‑Engines database and the most popular choice in the 2024 Stack Overflow Developer Survey, Postgres continues to solidify its position as the preferred database for developers and DBAs.

Postgres releases typically occur between September and October, with continuous updates that drive broad adoption. This article highlights the most impactful features of Postgres 17 that will further its adoption and make it an even more attractive option for applications.

Incremental Backup: Efficient Core Support

For years, Postgres users relied on external tools for incremental backups. PostgreSQL 17 now includes a native incremental backup capability that uses a fine‑grained, block‑level approach, providing a highly optimized processing pipeline. This core backup feature is stable from day one and will be available in all future major and minor releases.

The new functionality also offers a more efficient method to merge incremental backups offline into full backups, reducing server load typically required for full backups. Designed and implemented by EDB staff, veteran Postgres hacker Robert Haas, and other contributors, the feature is expected to be warmly received by DBAs managing on‑premises Postgres deployments.

Additional hooks allow storage to many different targets and support compression and tarball storage. The built‑in backup capability leads to faster recovery times, significantly shortening Recovery Time Objectives (RTO) for enterprises running large databases or high‑availability systems, making Postgres more appealing for organizations where downtime has a major business impact.

SQL/JSON and JSON Enhancements: Meeting Developers’ Needs

SQL/JSON is part of the SQL:2023 standard, offering developers a set of powerful functions to interact with JSON documents using familiar SQL syntax. While previous releases gradually improved SQL/JSON support, PostgreSQL 17 marks a major leap.

New functions such as JSON_TABLE(), JSON_EXISTS(), JSON_QUERY(), JSON_VALUE(), and others provide richer constructors for building JSON objects. The JSONPath implementation also gains several developer‑friendly features, further boosting productivity.

With these enhancements, Postgres becomes a stronger alternative to document databases for applications that store JSON documents, while still delivering unified transactional behavior, security models, indexing, and other database semantics across relational, spatial, vector, and document data types.