PostgreSQLhard

Why do we use connection pooling with Postgres (e.g., PgBouncer)?

Answer

Creating many Postgres connections is expensive (memory, process/thread work). A pool reuses connections and limits concurrency to protect the DB; you trade some session-level features (depending on pooling mode) for stability.

Related questions

PostgreSQL

Materialized view vs view: what’s the difference?

#postgres#views#materialized-view

PostgreSQL

PostgreSQL bloat: what causes it and how do you reduce it?

#postgres#mvcc#vacuum

PostgreSQL

PgBouncer pooling modes: why can transaction pooling break some apps?

#postgres#pgbouncer#pooling

PostgreSQL

PostgreSQL `COPY`: what is it used for and why is it fast?

#postgres#copy#bulk-load

PostgreSQL

What is a partial index in PostgreSQL and when is it useful?

#postgres#index#partial-index

PostgreSQL

What is `work_mem` and why can sorts or hash joins spill to disk?

#work_mem#sort#hash-join