PlanetScale Engineer Explains Kubernetes Feedback Loops by Running Postgres by Hand
bibryam · x · 2026-10-06
PlanetScale engineer Fatih Arslan published a deep-dive on why Kubernetes excels at running workloads at scale, framed through control theory.
Instead of starting with Kubernetes, he begins by operating a Postgres database by hand on a single Linux box, letting the feedback loop emerge naturally: an operator is a closed-loop feedback controller — the same loop as a thermostat or cruise control — just controlling a database.
Key points:
- Kubernetes self-heals by repeatedly comparing desired and observed state, not by replaying events;
- The piece builds up the production pieces step by step: a store, watches, queues, retries, idempotency, fan-in, and eventual consistency;
- It ends by examining what one of these loops looks like in a real operator.
His practical takeaway: operators move complexity into code — they don't remove it. Whether Kubernetes is worth it depends on whether scale and availability justify the machinery.
More from Infra
- Drax datacentre would burn 4.9m tonnes of wood a year, emissions near double Gatwick flights — nordicinst · 2026-10-06
- Singapore data center operator DayOne files for US IPO after H1 revenue tripled to $512M — zephyr_z9 · 2026-10-06
- Strata claims 6GB VRAM can match RTX 5090-level inference, with full Qwen4 support planned — lxfater · 2026-10-06
- ComfyUI benchmark: Flux 1 dev FP8 tested across 13 GPUs — Ok_Contribution8157 · 2026-10-06
- ODS offers one-click local AI: auto hardware detection, model setup, agents and plugins — Teknium · 2026-10-06
- One ENV Var Cut This Cloud Run Job's Billable Time by ~50%: NODE_COMPILE_CACHE — TechNadu · 2026-10-06