All you need is PostgreSQL
The article argues that PostgreSQL alone can handle many tasks typically requiring separate tools, such as queuing, caching, full-text search, and analytics. It presents PostgreSQL as a versatile, single-database solution for most application needs, reducing infrastructure complexity.
Background
- This is an essay arguing that PostgreSQL, a free open-source relational database, can handle workloads that many developers split across multiple specialized tools (Redis for caching, RabbitMQ for message queues, Elasticsearch for search, etc.).
- The author is a software engineer pushing back on the trend of "polyglot persistence" — using different databases for different job types — which increases complexity.
- Key context: The "PostgreSQL vs. many tools" debate has been active for years, with many developers rediscovering that Postgres extensions (e.g., LISTEN/NOTIFY for pub/sub, full-text search, JSONB for document storage, native partitioning) let one database do the work of several.
- The post comes from a personal blog, not a company, and reflects a growing sentiment in the backend engineering community that simpler architectures (fewer moving parts) often outperform fancy distributed stacks for most real-world projects.
- Why it matters: Teams often default to a "new tool for every new problem" pattern, but each extra service adds operational cost; this essay shows how a single database can eliminate that overhead for many common use cases.