AWS: Zero to Hero — Week 1: What Is "The Cloud," Actually?
Starting a new series: learning AWS from zero, one real topic a week, with Claude as my research and drafting partner. Week 1 breaks down what "the cloud" actually is: the problem it solves, public/private/hybrid deployment models, IaaS vs PaaS vs SaaS, why virtualization is the real unlock, and the economics behind why cloud (and now AI) keeps accelerating. Every post has a mandatory deep-dive plus an optional "my take" from me.
Series: AWS: Zero to Hero (9 articles)
- 1AWS: Zero to Hero — Week 1: What Is "The Cloud," Actually? This article
Starting a new series
I've never touched AWS before this month. Not once. So instead of pretending otherwise, here's the honest version of what this series is: every week, I'm going to actually learn one real piece of AWS and cloud computing, and write it up here as I go — partly so I don't lose the thread myself, and partly because if I'm confused about something as a beginner, chances are you are too.
I'm not doing this entirely solo, and I'd rather say that upfront than pretend otherwise — I'm using Claude (Anthropic's AI) as a research and drafting partner throughout this series: digging up accurate, current details, and helping structure what I've learned into something readable. The actual learning, reading, and hands-on practice is mine. I just don't see the point in hiding the tool that's helping me write faster.
Each week's post will have two parts:
- The content — the actual AWS topic for the week, explained properly. This is the mandatory part, every week, no matter what.
- My take — whatever I personally want to add: what confused me, what clicked, what I disagree with, or nothing at all. This part's optional — some weeks I'll have opinions, some weeks I'll just let the content speak for itself.
(Quick heads-up: my college midterms and end-sem exams land in the middle of this series in November — those weeks will be shorter, minimal-depth posts just to keep the streak alive. Every other week, expect the real depth.)
With that out of the way — Week 1. Before touching a single AWS service, I wanted to actually understand what "the cloud" is, because I realized I'd been using the word for years without a real definition behind it.
The Deep Dive
The problem cloud computing actually solves
Before "the cloud" was a marketing term, companies that needed computing power had exactly one option: buy physical servers, put them in a room (or pay someone else to), and maintain them yourself. That created a genuinely painful planning problem. Guess your future traffic too high, and you've sunk money into machines sitting idle most of the year. Guess too low, and the one week your product goes viral is the week your servers fall over — which is, ironically, the worst possible moment to be down.
This is the core problem cloud computing was built to solve: decoupling "how much computing power you need" from "how much hardware you own."
So what actually is "the cloud"?
Strip away the marketing, and cloud computing is just this: on-demand access to computing resources — servers, storage, databases, networking — delivered over the internet by a provider who owns and maintains the actual physical hardware, and who charges you based on what you actually use rather than what you own.
AWS (or Azure, or Google Cloud) owns enormous data centers full of real physical machines. What you're renting isn't a fixed slice of a specific machine — it's capacity, delivered virtually, that can grow or shrink with your actual demand. That distinction — capacity instead of hardware — is what makes everything else about cloud computing possible.
The three deployment models
Not all "cloud" looks the same. There are three flavors, and the difference actually matters when you're deciding how to run something:
- Public cloud — you're using shared infrastructure owned and operated by a third party (AWS, Azure, GCP). Multiple customers' workloads run on the same underlying hardware, kept securely isolated from each other. This is what most people mean when they say "the cloud," and it's what this whole series is about.
- Private cloud — infrastructure dedicated entirely to one organization, either self-hosted or hosted by a provider but never shared with anyone else. Common in banking, healthcare, and government, where compliance rules are strict about exactly where data physically lives.
- Hybrid cloud — a deliberate mix of both, usually because some workloads (sensitive data, legacy systems) stay on private infrastructure while everything else moves to public cloud for its flexibility and cost advantages.
A quick bonus layer: IaaS, PaaS, and SaaS
While I was reading about deployment models, I kept running into a second, related classification — how much of the stack the provider manages for you. Worth knowing early, because AWS service names will make a lot more sense once this clicks:
| Model | You manage | Provider manages | AWS example |
|---|---|---|---|
| IaaS (Infrastructure as a Service) | OS, apps, data | Physical hardware, virtualization, networking | EC2 |
| PaaS (Platform as a Service) | Just your application code | OS, runtime, scaling infrastructure | Elastic Beanstalk |
| SaaS (Software as a Service) | Nothing — you just use it | Literally everything | Amazon WorkMail |
Most of this series lives in the IaaS-and-below territory, because that's where you actually learn how things work rather than just consuming a finished product.
The real unlock: virtualization
Here's the part that actually explains how any of this is possible: virtualization. A single physical server can run software called a hypervisor, which carves that one machine into multiple isolated virtual machines — each with its own OS, each completely unaware the others exist, each unable to see or touch the others' data.
This is the actual technical foundation the entire cloud industry is built on. It's what lets AWS take one physical rack of servers and safely rent out slices of it to hundreds of different, mutually distrusting customers at once. Without safe virtualization, "cloud computing" as a shared, multi-tenant, pay-as-you-go model simply couldn't exist — you'd be back to everyone needing their own dedicated hardware. (AWS has evolved this a long way since EC2 launched in 2006 — there's a whole custom system called Nitro doing this job now, which is exactly what next week's post is about.)
Why this actually took off — the economics, not the hype
Here's the answer to the question that got me into this in the first place: why is cloud "getting popular" now, two decades after AWS launched?
The honest answer is it's not really new — it's a slow, compounding economic shift that's finally hit critical mass:
- Capex → Opex. Owning servers is a capital expense — a large amount of money spent upfront, whether you end up using the hardware fully or not. Cloud is an operating expense — you pay only for what you consume, when you consume it. For any company (especially a startup) with uncertain growth, that shift alone changes what's financially possible.
- Elasticity. Traffic spikes stop being a crisis. Infrastructure can scale up in minutes and back down just as fast, so you stop paying for capacity you're not using during quiet periods.
- Undifferentiated heavy lifting, removed. That's Amazon's own phrase for it, and it's a good one — running power, cooling, physical security, and hardware refreshes for a data center adds zero unique value to almost any product. Cloud lets companies stop doing that work themselves and spend that energy on what actually makes their product different.
- AI is the newest accelerant. Training and running modern AI models needs enormous, bursty amounts of specialized (and expensive) compute — exactly the kind of need nobody wants to buy permanently. This is a big part of why cloud adoption is accelerating again right now, and it's also exactly where this series is headed in a few weeks (Parts 11–13).
My Take
(Leaving this one for myself to fill in — some weeks I'll have something to add here, some weeks I might not.)
Next week — Part 2: AWS's global infrastructure, and what's actually happening underneath an EC2 instance when you launch one.
Series: AWS: Zero to Hero (9 articles)
- 1AWS: Zero to Hero — Week 1: What Is "The Cloud," Actually? This article
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article