Skip to content
Opensource internals
  • Posts

Performance

tmux Batches Terminal Output: One Write() Per Event Loop Iteration

Sep 5, 2026 10 min read Jay Hubbert

Every byte tmux sends to a terminal goes through one function that appends to a per-client buffer, and that buffer is flushed with a single write() per event loop iteration.

TmuxPerformanceInternalsTty

Read more →

PostgreSQL Computes Page Checksums Fast: Thirty-Two Parallel FNV-1a Hashes

Sep 5, 2026 29 min read Jay Hubbert

The page checksum runs 32 FNV-1a hashes in parallel over a 32-column view of the page to hide multiply latency — and 32 is baked into the on-disk format.

PostgresPerformanceStorageInternals

Read more →

PostgreSQL Deduplicates B-Tree Entries: Posting Lists and In-Place Splits

Sep 1, 2026 46 min read Jay Hubbert

B-tree leaves store one posting list tuple per distinct key instead of one entry per row, and an over-long list is split in place inside the insert’s critical section.

PostgresPerformanceBtreeIndexInternals

Read more →

PostgreSQL Does Async I/O with io_uring: A Closed-Loop Read-Ahead Controller

Aug 27, 2026 32 min read Jay Hubbert

PostgreSQL 20’s AIO subsystem has an io_uring backend plus a read-ahead controller that doubles its look-ahead distance only when the backend actually had to wait for I/O.

PostgresPerformanceIoInternals

Read more →

tmux Redraws Only What Changed: Five Layers of Less Work

Aug 21, 2026 11 min read Jay Hubbert

tmux never redraws ’the screen’ — it redraws the difference.

TmuxPerformanceInternalsRedraw

Read more →

← Newer Page 20 of 23 Older →

© 2026 Jay Hubbert · Built with Hugo 0.168.0

RSS ·