Kubernetes: Scaling PostgreSQL to Power 800 Million ChatGPT Users | Kaptain

🔗 View in your browser. | ✍️ Publish on FAUN.dev | 🦄 Become a sponsor

Kaptain

#Kubernetes #Docker #DistributedSystems

🔍 Inside this Issue

Cost-cutting pragmatism is colliding with scale: teams are trimming observability bills and self-hosting Docker while Postgres quietly powers hundreds of millions of users. With Docker drifting toward AI, Kubernetes loosening storage handcuffs, and Go devs using AI they still do not trust, the details below are where the real lessons hide.

🧱 Building Production-Grade Micro services on Azure Kubernetes
⚓ Dockhand - The Ultimate Self-Hosted Docker Management Tool
📊 Go Developer Survey Is Out: What 5,379 Go Developers Actually Want Next
🐘 Scaling PostgreSQL to power 800 million ChatGPT users
☸️ v1.35: Mutable PersistentVolume Node Affinity (alpha)
🐳 What has Docker become?

Ship smarter this week.

See you in the next issue!
FAUN.dev() Team

ℹ️ News, Updates & Announcements

faun.dev

Go Developer Survey Is Out: What 5,379 Go Developers Actually Want Next

The 2025 Go Developer Survey is in, and AI’s gaining ground. Devs are leaning on it for unit test scaffolding, autocompletion, and similar grunt work. But trust? Still shaky. Quality’s not there yet.

Elsewhere, they’re bumping into old problems: core tools still hard to navigate, non-idiomatic code patterns still causing trouble, and a rising call for clearer best practices and better trust signals in modules.

👉 Enjoyed this?Read more news on FAUN.dev/news

🔗 Stories, Tutorials & Articles

noted.lol

Dockhand - The Ultimate Self-Hosted Docker Management Tool

Dockhand just dropped, and it's aiming straight at the bloated SaaS stack.

It’s a fully self-hosted Docker management tool with zero license walls. Local or remote? Doesn’t matter. It even plays nice behind NAT using outbound WebSocket agents.

You get container lifecycle controls, a visual Compose editor, real-time logs, vuln scans, and terminal access, wrapped in a clean UI. Data lives in SQLite or Postgres, your call.

kubernetes.io

v1.35: Mutable PersistentVolume Node Affinity (alpha)

Kubernetes 1.35 (alpha) cracks open PersistentVolume node affinity. You can now update it on the fly.

Before, it was locked down - once set, it stayed set. That got in the way of shifting workloads when disks were upgraded or moved across zones. Now? More flexibility. Less pain.

tuananh.net

What has Docker become?

Docker’s not just about containers anymore. It’s pivoting hard into AI infrastructure - with some teeth.

The new Model Runner, GPU offloading, and fresh AI-native integrations with Google Cloud and Vercel show where it’s headed: less dev environment, more AI runtime engine.

Under the hood, Docker dropped over 1,000 Docker Hardened Images into the Apache 2.0 open source bin - heavy counterweight to Chainguard’s momentum in secure images.

openai.com

Scaling PostgreSQL to power 800 million ChatGPT users ^✅

OpenAI pushed PostgreSQL to handle millions of QPS across 800M users. How? Nearly 50 read replicas, heavy read offloading, and serious trimming on write pressure.

Writes? Sent elsewhere. Sharded systems like CosmosDB, lazy writes, and app-level tweaks helped sidestep PostgreSQL’s MVCC write amplification mess.

Cache misses don’t get a free pass either - a custom cache locking setup rate-limits bursty traffic before it hits the primary.

Still not enough? They’re testing WAL relay replication. Relay nodes forward the write-ahead log, offloading replicas and buying time beyond normal scaling ceilings.

The bigger picture: With the right hacks - sharding, caching, WAL relays - PostgreSQL can play at global scale.

medium.com

Building Production-Grade Micro services on Azure Kubernetes

A team running microservices on Azure Kubernetes Service gave their setup a smart overhaul: critical state stayed managed in PostgreSQL, but compute and observability went DIY. The payoff? Major cost cuts. Interrupt-friendly jobs landed on spot instances, and they ditched pricey per-GB logging for a homegrown solution.

👉 Got something to share? Create your FAUN Page and start publishing your blog posts, tools, and updates. Grow your audience, and get discovered by the developer community.

⚙️ Tools, Apps & Software

github.com

ArmDeveloperEcosystem/kubearchinspect

Check if container images in a Kubernetes cluster have arm architecture support.

github.com

kubernetes/node-problem-detector

This is a place for various problem detectors running on the Kubernetes nodes.

github.com

agentkube/agentkube

Agentkube is an AI-powered Kubernetes management platform that bridges the gap between developers and complex cluster operations.

github.com

luxury-yacht/app

Luxury Yacht - Sail the seas of Kubernetes in style!

github.com

stern/stern

Multi pod and container log tailing for Kubernetes

👉 Spread the word and help developers find and follow your Open Source project by promoting it on FAUN. Get in touch for more information.

🤔 Did you know?

Did you know that in Kubernetes a container’s preStop hook runs before the process ever gets a SIGTERM? The kubelet runs the preStop hook and waits up to the Pod’s terminationGracePeriodSeconds before sending SIGTERM, so if the hook uses up the grace period the process can go from hook straight to SIGKILL. This means traffic draining on SIGTERM won’t happen during preStop unless the hook itself does it.

🤖 Once, SenseiOne Said

"Stateless containers are easy; stateful guarantees are where distributed systems start billing you. Kubernetes will schedule your pods; it won't schedule your consistency."
SenseiOne

(*) SenseiOne is FAUN.dev’s work-in-progress AI agent

⚡Growth Notes

Most Kubernetes folks eventually stall at writing more YAML, but the ones who quietly pull ahead routinely trace problems starting from a user symptom down through ingress, service, endpoints, kube-proxy/ CNI, and into pod logs, treating that path as something they should be able to walk from memory. Make it a habit to reconstruct that full request path in new clusters you join, because teams subtly start routing the gnarlier production incidents to the person who can do this without guesswork, and the opportunities tend to follow the incidents.

Each week, we share a practical move to grow faster and work smarter

👤 This Week's Human

This week, we’re highlighting Yusuf Aytaş, Senior Engineering Leader at Workday, who has led SRE, data science and engineering, backend, and platform engineering teams across EMEA, APAC, and North America. He writes the Software Engineering Handbook and lectures at Dublin Business School, distilling lessons from running systems on AWS, GCP, Azure, Kubernetes, Kafka, Spark, PostgreSQL, and Cassandra.

💡 Engage with FAUN.dev on LinkedIn — like, comment on, or share any of our posts on LinkedIn — you might be our next “This Week’s Human”!

😂 Meme of the week

❤️ Thanks for reading

👋 Keep in touch and follow us on social media:
- 💼LinkedIn
- 📝Medium
- 🐦Twitter
- 👥Facebook
- 📰Reddit
- 📸Instagram

👌 Was this newsletter helpful?
We'd really appreciate it if you could forward it to your friends!

🙏 Never miss an issue!
To receive our future emails in your inbox, don't forget to add community@faun.dev to your contacts.

🤩 Want to sponsor our newsletter?
Reach out to us at sponsors@faun.dev and we'll get back to you as soon as possible.

Kaptain #513: Kubernetes: Scaling PostgreSQL to Power 800 Million ChatGPT Users
Legend: ✅ = Editor's Choice / ♻️ = Old but Gold / ⭐ = Promoted / 🔰 = Beginner Friendly

You received this email because you are subscribed to FAUN.dev.
We (🐾) help developers (👣) learn and grow by keeping them up with what matters.

You can manage your subscription options here (recommended) or use the old way here (legacy). If you have any problem, read this or reply to this email.