Kubernetes

This post walks through deploying the full DeepSeek-r1-671B model on Kubernetes with SGLang for production-grade, multi-node GPU inference. It explains how to orchestrate elastic multi-GPU workloads using LeaderWorkerSet and Volcano, optimize performance via RadixAttention and KV cache reuse, integrate Prometheus/Grafana for SLA-grade monitoring, and contrasts this K8s+SGLang stack with Ollama. A step-by-step environment and YAML guide is included.

This post walks through using Thanos to aggregate metrics from multiple Prometheus instances for high availability, centralized querying, and long‑term retention in object storage. It explains core components like Sidecar, Store, Query, and Ruler, and shows key Kubernetes deployment details, including image mirror replacement, exposing NodePort Services, and handling NetworkPolicy pitfalls.

Step-by-step guide to building a Kubernetes cluster with kubeadm on three Ubuntu 20.04 servers. Covers environment preparation, disabling SWAP and firewall, kernel and containerd configuration, installing kubeadm/kubelet/kubectl, initializing the control plane, deploying Flannel CNI, enabling shell completion, and joining worker nodes, with ready-to-use scripts and version notes.

A structured guide to essential kubectl commands, covering creation, querying, deployment, scaling, and troubleshooting. Explains key subcommands like apply, rollout, logs, and exec, and includes cluster management utilities plus a quick reference to common Kubernetes resource names and abbreviations to boost your daily Kubernetes efficiency.

Step-by-step guide to deploying applications on Kubernetes, from using `kubectl run` to managing Pod and Deployment YAML. Learn how labels connect Deployments and Pods, how to scale replicas, forward ports, and roll back versions. Includes a concise cheatsheet of common `kubectl` commands for querying resources, inspecting logs, and troubleshooting in real clusters.

This tutorial walks you through deploying a Kubernetes cluster with Minikube on Debian 11, using either a Hetzner dedicated server or a typical VPS. It covers environment and driver requirements, installing and verifying kubectl and Minikube, starting Minikube with the Docker driver under root (and handling the permission warning), checking cluster state, and enabling remote access to the Kubernetes Dashboard.

This post traces the evolution from traditional deployment to containers, then introduces Kubernetes as a container orchestration platform. It explains K8s origins, core features, and Master/Node components, and walks through deploying Nginx to illustrate scheduling, self‑healing, autoscaling, and service discovery, helping you get started quickly.