Programming, architecture, application development, message queues, middleware, databases, containerization, big data, image processing, machine learning, AI, personal growth.
The author chronicles the challenges and solutions of running large language models locally using Ollama, experimenting with cloud GPUs on Google Colab, managing Python dependencies through Docker, and ultimately fine‑tuning a small Qwen model, providing a practical guide for AI enthusiasts.
This article provides a comprehensive overview of Ollama, a lightweight tool for running large language models, detailing its client‑server architecture, local storage layout, and the step‑by‑step workflow of user interactions with the model.
This article explains how Kubernetes leverages the Linux kernel security mechanisms Seccomp, AppArmor, and SELinux together with Pod Security Standards and the built‑in admission controller to enforce fine‑grained security policies for container workloads in cloud‑native environments.
This article explains how kube-proxy and CNI plugins cooperate within a Kubernetes cluster to translate Service ClusterIP requests into real Pod IPs, detailing the Netfilter, iptables/ipvs rule configuration, overlay networking modes, and the emerging proxy‑free approaches using eBPF.
This article explains the architecture, components, and operational flow of a CSI driver using NFS as a concrete example, covering CSI fundamentals, sidecar containers, dynamic volume provisioning, pod creation, and the role of the Linux VFS in exposing remote storage to applications.
This article explains the Open Container Initiative (OCI) standards—Image-spec, Distribution-spec, and Runtime-spec—detailing how code is built into OCI images, pushed to registries, and executed via Kubernetes components such as kubelet, containerd/CRI‑O, and runc or kata‑containers, ensuring portable, interoperable containers.
This article explains how the kubectl command interacts with the Kubernetes control plane, detailing one‑way HTTPS requests and two‑way WebSocket communications, and walks through the full execution flow of a kubectl exec command across kube‑api‑server, kubelet, and the CRI runtime.
This article explains the fundamentals of DNS, its role as an application‑layer protocol, and how Kubernetes generates and manages each pod's /etc/resolv.conf through kubelet, covering nameserver, search, ndots options, dnsPolicy choices, and custom dnsConfig examples.
This article explains Kubernetes' container networking fundamentals, covering the basic network requirements, IPAM implementation with node CIDR allocation, Linux veth and bridge mechanisms, overlay solutions like VXLAN, and the role of BGP for cross‑cluster pod reachability.
When a new Pod is created, Kubelet coordinates with the CRI and CNI components to set up the sandbox, configure networking, pull images, create and start containers, using gRPC calls and command‑line interactions, with details varying across container runtimes such as containerd, cri‑o, and Docker.
