🗣️ @garitweets, Product Manager at @googlecloud, discusses the evolution of Kubernetes cluster scaling. He explains how Karpenter achieved 40% cost savings through workload consolidation Watch: https://t.co/3Tccfesbkw

This week on the Learn Kubernetes Weekly: ⚙️ gRPC with ALB and Traefik 🧭 Prevent Failures with Topology Spread 📜 Demystifying Kubernetes YAML 🔗 Shared Socket with eBPF 🌐 Kubernetes Networking Guide ⭐️ @testkubeio Read it now: https://t.co/i5beD2n65l

This tutorial walks through Kubernetes networking theory and hands-on examples, covering pod connectivity, services, DNS, load balancing, network policies, and common pitfalls ➜

🗣️ @_enfinity walks through his project of building a lightweight version of Kubernetes from scratch in Go https://t.co/QAzAECg0Xn 🌟 @StormForgeIO 🎙 @birthmarkbart

Kubero is a fully self-hosted Internal Developer Platform that brings Heroku's workflows to your Kubernetes cluster It enables you to deploy your applications with a few clicks, has a built-in CI/CD pipeline, and supports multiple staging environments ➜

This tool runs a Kubernetes operator that manages OpenTelemetry Collectors and auto-instrumentation for your workloads ➜

This article shows how to use Skaffold and local dev workflows to shorten the feedback loop when coding for Kubernetes, enabling you to build, deploy and test changes continuously on a local cluster ➤

https://t.co/N3aCeG2Hl4

KAI Scheduler is a Kubernetes-native scheduler optimized for large-scale AI/ML workloads It supports batch scheduling, hierarchical queues, GPU sharing, and dynamic resource allocation to maximize utilization and fairness across tenants ➤

This article explains how Kubernetes integrates image compatibility metadata via Node Feature Discovery (NFD), enabling workloads to specify and validate required host OS, kernel modules, and hardware features before scheduling ➤ https://t.co/JtSCBu9Ylv

🗣️ Sai Vennam, Principal Specialist Solution Architect, Container @ @awscloud, shares three emerging Kubernetes tools: Karpenter, Kubeflow and Ray, and Kwok Watch the full interview: https://t.co/rB9pSEhBbx

This plugin extends the NVIDIA Device Plugin to let Kubernetes pods scheduled by Volcano share GPUs as vGPUs while ensuring hard isolation of resources using HAMi-core or dynamic-MIG modes ➤ https://t.co/VVI9AlZxAo

This article explains the three core parts of a Kubernetes YAML file: - metadata, - spec, - and status It shows how your YAML declares desired state and how Kubernetes uses that with its reconciliation loop ➜

This article walks through how to route gRPC traffic via an AWS ALB into Traefik as the Kubernetes Ingress, and explains why health-checks and protocol mismatches (HTTP/2 vs gRPC) can break things ➜

This article explains how to build a custom event aggregation system for Kubernetes that groups, correlates, and retains events beyond default capacity ➤

k8s-doctor is a kubectl cluster triage plugin that helps you diagnose cluster issues — like `brew doctor` but for Kubernetes ➜

💰 Here's something that surprised me: most Kubernetes cost waste comes from CPU, not memory. I asked Rafa Brito (who specializes in Kubernetes cost optimization) what the data shows across their customer base: cost reductions between 35-60%, averaging around 54%. But here's

This article explains how Pod Topology Spread Constraints can mis-distribute pods during rolling updates, and shows how using the matchLabelKeys field (like `pod-template-hash`) lets Kubernetes keep pods evenly spread even across updates ➜

This week on the Learn Kubernetes Weekly: 🔥 AI Infrastructure on K8s 🏠 HA Databases on K8s at Airbnb ⚡ Faster Node & Pod Startup 🎯 Pod Priority and Preemption ⚖️ Cost vs Resilience in Scheduling ⭐️ @testkubeio Read it now: https://t.co/Ot8jacS0mU

Reloader is a Kubernetes controller that monitors changes in ConfigMap and Secrets and triggers rolling upgrades on Pods with their associated Deployment, StatefulSet or DaemonSet ➜