PostgreSQL is a powerful, open-source object-relational database system with nearly 40 years of active development. It has earned a strong reputation for reliability, data integrity, and an extensive feature set. It is ACID-compliant, runs on all major operating systems, and is the preferred open-source relational database for organisations of all sizes. PostgreSQL 18 — released on 25 September 2025 — is the current stable version. Official documentation: https://www.postgresql.org/docs/ and https://www.postgresql.org/about/
Core Capabilities
PostgreSQL supports an extensive range of data types beyond the standard relational primitives. In addition to integers, strings, booleans, and date/time types, it natively supports arrays, ranges, JSON and JSONB documents, XML, key-value stores (hstore), and geospatial geometry types via the PostGIS extension.
As of PostgreSQL 18, the database conforms to at least 170 of the 177 mandatory features for SQL:2023 Core conformance — more than any other relational database available today. This level of SQL standard conformance means that queries, constraints, and schemas written to standard SQL work as expected without proprietary syntax.
PostgreSQL supports full ACID transactions, foreign keys, joins, views, triggers, stored procedures, and complex index types including B-tree, Hash, GiST, GIN, and BRIN. It is highly extensible — many features, including index types, have defined APIs that allow the database to be extended without modifying the core.
- ACID-compliant since 2001 — data integrity is guaranteed at the transaction level
- Rich data types: JSON/JSONB, arrays, ranges, UUID, geospatial via PostGIS
- SQL:2023 Core conformance — 170 of 177 mandatory features
- Extensible: custom index types, functions, data types, and extensions
- Supports complex index types: B-tree, Hash, GiST, GIN, BRIN
PostgreSQL 18: What Changed
PostgreSQL 18 introduces several significant improvements. The headline feature is a new asynchronous I/O (AIO) subsystem that enables the database to queue multiple read requests simultaneously. Benchmarks from the PostgreSQL Global Development Group show up to 3× performance improvements for read-heavy workloads, particularly benefiting sequential scans, bitmap heap scans, and vacuum operations.
Version 18 also adds native support for UUID version 7 (uuidv7()), which generates timestamp-ordered UUIDs. This resolves a long-standing performance debate about primary key strategies — UUIDv7 provides the global uniqueness of UUIDs alongside the sequential ordering that makes B-tree indexes efficient.
Other notable additions in version 18 include: virtual generated columns that compute values at query time (now the default for generated columns); support for B-tree skip scans that allow multi-column indexes to be used even when leading columns are not specified; OAuth 2.0 authentication support for SSO integration; and the ability to retain planner statistics through major version upgrades, reducing query performance degradation after an upgrade.
Reference: https://www.postgresql.org/docs/release/18.0/
- Asynchronous I/O subsystem: up to 3× read performance improvement
- Native uuidv7() function for sequential, globally unique primary keys
- Virtual generated columns computed at query time by default
- B-tree skip scan: use multi-column indexes without specifying all leading columns
- OAuth 2.0 authentication support for SSO systems
- Planner statistics retained through major version upgrades
Why We Use PostgreSQL as Our Default Database
PostgreSQL is our standard database for relational workloads. The decision is based on reliability, ecosystem maturity, and the absence of licensing risk. PostgreSQL uses the PostgreSQL License, a BSD-like permissive license that places no restrictions on commercial use — there is no risk of vendor lock-in or licensing changes.
On AWS, PostgreSQL is available as a managed service through Amazon RDS and Amazon Aurora (PostgreSQL-compatible), which handles backups, automated failover, read replicas, and patching. For products where operational overhead matters, this managed path significantly reduces the database administration burden without giving up PostgreSQL's capabilities.
The PostgreSQL community maintains detailed documentation at https://www.postgresql.org/docs/ — widely considered one of the most thorough database documentation sets available. This makes onboarding new developers and resolving operational questions significantly faster than proprietary alternatives.