PostgreSQL 12, announced on October 3, 2019, gives application developers new ways to work with JSON data and generated columns, while changing how some common table expression queries are planned. For database teams, the practical promise is less friction in particular workloads—not a blanket speed-up. [shared-03ed6acce83a0bacef917af69439e5eb384487f7733b6a8e76f1e7e88de9f19a]
The project reported that its B-tree benchmark showed average space use falling 40% alongside an overall query-performance gain. That result belongs to the stated benchmark, not every business database. Partitioned-table improvements likewise matter most for specific patterns, such as queries touching a small share of tables with thousands of partitions. A WITH query is eligible for automatic inlining only when it is nonrecursive, side-effect-free and referenced once. [shared-03ed6acce83a0bacef917af69439e5eb384487f7733b6a8e76f1e7e88de9f19a]
There are concrete operational changes too: REINDEX CONCURRENTLY can rebuild indexes without blocking writes, while checksum settings can now be changed with pg_checksums—but only with the cluster offline. JIT compilation is enabled by default only in installations built or packaged with LLVM. These qualifications shift work rather than erase it: teams gain options, but must match them to their systems and schedule downtime where required. The release announcement supplies no independent workload data to show how much time typical operators will save. [shared-03ed6acce83a0bacef917af69439e5eb384487f7733b6a8e76f1e7e88de9f19a]