Takeaways

• Async I/O cuts cold-cache read latency materially on NVMe.
• Skip-scan makes more multicolumn indexes usable without rewrites.
• Upgrade path is smooth; test the new planner stats first.

Highlights

• Async I/O subsystem (effective_io_concurrency now matters more).
• B-tree skip scan for leading-column gaps.
• Better partition-wise joins.

Summary

Postgres 18 is an incremental but meaningful release: the headline is the new asynchronous I/O path, with skip-scan and planner improvements close behind. For most homelabs the upgrade is low-risk and worth it for the read-latency wins.