About this article
Thank you for visiting this site. This article is a guide listing all 12 articles in the “System Architecture” category of the Architecture Crash Course for the Generative-AI Era series.
System architecture is the layer that decides the infrastructure skeleton — hardware, OS, networking, and cloud. If you don’t decide “where, on what, and how to run things” before writing code, the cost of fixing it later multiplies by tens.
This category covers every decision axis for system foundations, from choosing a cloud vendor to managing costs.
Article index
1. System Architecture Overview — The Skeleton You Decide First
A bird’s-eye view of what system architecture is and why it must be decided first. Covers configuration patterns by scale and AI-era selection criteria — a map for everything that follows. Read this first if you’re entering this category.
2. Choosing an Application Form — Web / Native / Hybrid
Compares native, web (SaaS/PaaS/IaaS/FaaS), and hybrid forms, and derives selection criteria from target users and operational capacity. Useful when you’re stuck on the first fork: “should our service be a web app or a native app?“
3. Choosing a Deployment Model — On-Prem / Cloud / Hybrid
Compares five forms: on-premises, public cloud, private cloud, hybrid, and multi-cloud. Shows recommended configurations by scale and industry — useful as evidence when getting deployment decisions approved.
4. Choosing a Cloud Vendor — AWS / Azure / GCP
Compares the big three — AWS, Azure, and Google Cloud — on global market share, key services, recommended vendor by scale, and how to think about lock-in. Answers the most common real-world question: “which cloud should we pick?“
5. Choosing a Runtime — VM / Container / Serverless / Wasm
Compares five layers: bare metal, VM, container, serverless (FaaS), and WebAssembly. Shows recommended configurations by use case. Gives you the selection criteria to avoid runtime mistakes — Docker, Kubernetes, Edge Runtime, and more.
6. OS Selection — Linux / Windows / UNIX and ARM
Your server OS is a 10-year relationship. Covers Linux distribution selection, EOL management, x86/ARM architecture, licensing models, and patching — the unglamorous but fatal-if-wrong fundamentals of OS selection.
7. Datastore Placement — RDBMS / NoSQL / Cache / Search Combinations
Explains how to combine RDBMS, NoSQL, cache, search engines, and object storage as a system-level concern. Also covers Polyglot Persistence and the minefield of schema changes.
8. Network Design Fundamentals — VPC / Subnet / CIDR
Cloud network design is a floor plan you can’t redraw later. Covers CIDR design, three-tier subnet structure, multi-AZ, external connectivity, and security groups — the proven patterns and the classic failure modes.
9. Security Foundations — Defense in Depth and Least Privilege
A map of the security foundations that must be embedded at the system architecture stage. Covers defense in depth, WAF/DDoS/IDS/IPS, IAM, encryption, and compliance — a security roadmap baked into the design.
10. Monitoring and Operations — The Three Pillars of Observability and the Four Golden Signals
Covers the three pillars of observability (Metrics/Logs/Traces) and the four golden signals, from monitoring platform selection to phased rollout plans by scale. The article that prevents the hellscape of “monitoring can wait.”
11. BCP/DR Design — RPO, RTO, and the 3-2-1 Rule
Covers how to set RPO/RTO, the availability staircase, four DR strategy patterns, the 3-2-1 backup rule, ransomware countermeasures, and recovery drills — the design for systems that don’t stop when failures or security incidents hit.
12. Cloud Cost Management (FinOps) — Guard with Design, Polish with Operations
Explains why cloud bills balloon and how to fight back with the three phases of FinOps. Covers pricing models, tagging strategy, management levels by scale, and the new cost dimensions introduced by LLMs — the art of managing costs from both design and operations.
Summary
This article listed all 12 articles in the System Architecture category of the Architecture Crash Course for the Generative-AI Era series.
System architecture is “the skeleton you decide first” — and the domain where changes later are the most expensive. After getting the big picture from the overview article, we recommend diving into the themes most relevant to your project.
For the overall series structure and other categories, see the master series index.
Hope you’ll check out the next article as well.