Explained

Short, clear write-ups on protocols, tools, and concepts. Each post takes one idea and explains it properly: what problem it solves, how it works, and where you will encounter it in the real world.

Posts are grouped by topic. No prior knowledge of the specific subject is assumed, just a working background as an engineer.

Networking

• Explained: STUN Protocol
  How peers behind NAT discover each other before a WebRTC call connects.
• Explained: NAT
  Why private IPs work on the internet and how routers translate addresses.
• Explained: DERP Protocol
  A relay that forwards encrypted packets when a direct peer connection is not possible.
• Explained: IPSec VPN Tunnels
  How IPSec negotiates, wraps, and protects packets at every hop.
• Explained: Webhooks
  How apps push real-time data updates without polling, and best practices for doing it securely.
• Explained: MQTT
  The lightweight pub/sub protocol behind most IoT communication, covering architecture and use cases.

Kubernetes

• Explained: Node Affinity
  Controlling which nodes a pod can and cannot be scheduled onto.
• Explained: Taints and Tolerations
  Preventing pods from landing on the wrong nodes, and granting exceptions when you need them.
• Explained: CoreDNS
  How Kubernetes resolves service names inside the cluster, and how to debug DNS failures methodically.
• Explained: Kubernetes Operators
  Encoding domain-specific operational knowledge as code that runs inside the cluster.

Distributed Systems

• Explained: Gossip Protocol
  How nodes in a cluster spread information to each other without a central coordinator.
• Explained: Bloom Filters
  A probabilistic data structure that answers "have we seen this?" using very little memory.
• Explained: Bastion Server
  A single hardened entry point that keeps your private infrastructure off the public internet.
• Explained: Distributed Storage Systems
  Architecture, replication, and consistency models behind scalable distributed storage.

Messaging and Streaming

• Explained: Kafka vs AMQP
  How the two messaging models differ in architecture, and when to reach for each one.

Data and Architecture

• Explained: Data Warehouse, Lake and Lakehouse
  The differences between the main data storage paradigms and how to choose between them.
• Explained: Lambda Architecture
  A data-processing pattern that combines batch and real-time layers to handle large-scale streams.
• Explained: Data Mesh
  A decentralized approach to data ownership where domain teams manage their own data as a product.

Enterprise Architecture

Part of the Enterprise Architecture series, which follows one fictional bank from first principles through to where the discipline stands today.

• Explained: Enterprise Architecture
  What the discipline actually is, and why whole-organization questions have no owner without it.
• Explained: The Four Architecture Domains
  Business, data, application, and technology: the layered model the entire discipline hangs off.
• Explained: The Zachman Framework
  A classification schema rather than a method, and how it contrasts with TOGAF.
• Explained: Choosing a Framework
  How much architecture rigor an organization actually needs, and why the framework question is the wrong one.
• Explained: Capability Mapping
  The single most useful EA artifact, and how it differs from an org chart and a process map.
• Explained: Architecture Governance
  Review boards, standards, and the constant tension between control and delivery speed.
• Explained: Conway's Law and EA
  Why systems come to mirror the organizations that build them, and what that means for architects.
• Explained: EA in the Age of Cloud and Microservices
  How self-service infrastructure destroyed central architecture's authority, and what replaced it.

Mainframe

• Explained: COBOL
  The language that still runs the world's most critical transactions, and why it has lasted this long.
• Explained: JCL
  How batch jobs are defined, submitted, and controlled on z/OS using Job Control Language.
• Explained: RACF
  The security and access control system that protects z/OS resources at every level.

Tooling

• Explained: WSL (Windows Subsystem for Linux)
  How Microsoft built native Linux binary support into Windows, without a VM or dual boot.
• Explained: WebDrivers
  How Selenium's WebDrivers interact with browsers to automate testing at the protocol level.

Posts are added to this page as they are published. If you are looking for a place to start, Explained: NAT is a good entry point.

Written by Phee Jay.