LearnKube

This article explains how to use ephemeral infrastructure where components are designed to exist only as long as needed, then disappear, making systems more stable It details how Kubernetes Pods are designed to be killed and replaced at any moment ➤ ku.bz/Wzhwfprrr

This case study explains how the authors used dynamic MIG partitioning to split large GPUs like NVIDIA A100/H100 into multiple isolated slices, letting many small jobs share GPU efficiently ➜ ku.bz/h4B9DHKH0

AWS at Booth 700: a mini theater with 20+ lightning talks, demo stations for AI and EKS, and a Sketch Lab that turns your drawings into custom stickers Visit Booth 700: ku.bz/bKNHfprg1 Full interview: ku.bz/_h47WT52q

This case study explains how Pinterest migrated its search system (Manas) to Kubernetes and tracked down a rare performance issue They found cAdvisor’s memory metric scanning caused huge latency spikes ➤ ku.bz/BS18f9fpM

The ACK Service Controller for Amazon RDS is a Kubernetes controller that enables you to manage Amazon RDS database instances directly from Kubernetes ➤ ku.bz/Bjlzbx970

Purr is a TUI wrapper for kubectl that provides completions for commands, resource caching with watch-based updates, persistent command history with fuzzy search, and an interactive file picker ➤ ku.bz/nD3c3q9Ml

lazydocker is a terminal UI for managing Docker containers and services, with log and metric graph viewing, container attachment, and execution of common Docker commands ➜ ku.bz/x2rRKH5n_

Kubex at Booth 589: GPU demos, a Wednesday DevEx talk, and raffles for a Steam Deck and Nintendo Switch GPU demo: ku.bz/mDhF0LKj_ Talk details: ku.bz/BT37R74fk Full interview: ku.bz/V-8twLbjh

Sloth generates Prometheus Service Level Objectives with reliable SLI recording rules and multi-window, multi-burn-rate alerts from simple YAML specs ➜ ku.bz/SLVjZfYbV

This article explains how to reduce a Kubernetes sidecar container from 421MB to 90MB by building a statically linked Go binary and using FROM scratch base image instead of Alpine Linux ➤ ku.bz/71cCMTf9S

NVIDIA Dynamo is a datacenter-scale distributed LLM inference framework supporting disaggregated prefill/decode, KV-aware routing, and dynamic GPU scheduling across vLLM, SGLang, and TensorRT-LLM ➜ ku.bz/3TsFWKPgK

This article explains how Dream11 built an in-house observability platform using SigNoz, ClickHouse, and OpenTelemetry to handle millions of metrics and traces across thousands of EC2 instances, saving millions in commercial tooling costs ➜ ku.bz/ZSb36t4t5

🗣️ Yue Yin, Software Engineer at ByteDance, discusses their Gödel scheduler processing 2,000 pods per second with a single shard Watch the full episode: ku.bz/lMpNng_33

This week on the Learn Kubernetes Weekly: 💰 Cost-Aware Scheduling 📐 Scaling Workloads with VPA 🕸️ Service Mesh Patterns 🐛 ZFS ARC Container Initialization Slowness 🍓 Developing on Raspberry Pi ⭐️ vCluster Read it now: kube.today/issues/175

What is Spectro Cloud bringing to KubeCon Europe 2026? Anthony Newman previews Booth 420: AI factories, VMware-to-KubeVirt migration, air-gapped ops for defense and public sector, and a Meta glasses giveaway ku.bz/35dSbG92z Full interview: ku.bz/8-BCyLWJD

This tutorial teaches how to build a production-like service mesh lab using Istio with a 3-tier application (Next.js, Go, Flask) on a local Kind cluster ➜ ku.bz/2kz6TNYpR

KubeGUI is a desktop application that provides a graphical interface for managing Kubernetes clusters with resource visualization and management capabilities ➤ ku.bz/sQltn0tyP

🗣️ Ron Matsliah from Next Insurance built an AI assistant that cut build debugging time by 75% — combining deterministic rules with AI, delivered straight into Slack ku.bz/PDdYfC00w 🌟 LearnKube 🎙 🎙Bart