PostgreSQL 16: the open source database that never stops innovating

PostgreSQL 16 introduces logical replication from standby, FULL OUTER JOIN parallelism, pg_stat_io for I/O analysis and SIMD acceleration. Nearly thirty years of continuous evolution.

Open SourceWeb Open SourcePostgreSQLDatabaseSQLPerformance

Nearly thirty years of evolution

PostgreSQL released its first public version in 1996. Since then, every major release has added features that in many cases anticipated or matched those of commercial databases. Version 16, released in September 2023, continues this tradition with improvements spanning replication, parallelism, observability and low-level performance.

Logical replication from standby

One of the most significant additions is the ability to perform logical replication from a standby server. In previous versions, only the primary server could publish logical changes to subscribers. This forced all replication load onto the primary, adding overhead to a system already under pressure in high-traffic environments.

With PostgreSQL 16, a server in streaming replication (physical standby) can act as a publisher for logical replication. The result is workload separation: the primary handles writes, the standby handles logical data distribution to other nodes, data warehouses or analytical systems.

Parallelism and FULL OUTER JOIN

The PostgreSQL 16 query planner extends parallelism support for FULL OUTER JOIN and RIGHT OUTER JOIN operations. In previous versions, these operations were executed on a single worker even when the planner could have distributed them across multiple cores. The improvement is particularly relevant for analytical queries on large tables, where complex joins often represent the bottleneck.

pg_stat_io and observability

The new pg_stat_io view provides detailed statistics on I/O operations: reads, writes, file extensions, broken down by backend type (client, autovacuum, checkpointer, WAL writer). Before pg_stat_io, analysing PostgreSQL’s I/O behaviour required external tools at the operating system level. The integrated view allows correlating I/O patterns directly with database activity.

SIMD acceleration

PostgreSQL 16 introduces the use of SIMD (Single Instruction, Multiple Data) instructions to accelerate frequent internal operations, particularly string scanning for JSON and ASCII parsing. The implementation leverages vectorised instructions available on modern CPUs to process more bytes per clock cycle, with measurable throughput improvements in text processing and semi-structured data functions. The database is released under the PostgreSQL License, a permissive licence similar to BSD.

Link: postgresql.org

Need support? Under attack? Service Status
Need support? Under attack? Service Status