Skip to content
Tutorials Center
My learning

Topics

Lesson 1 of 45 ยท Cloud Basics

What Is Cloud Computing?

Learn what cloud computing is: on-demand resources, pay-as-you-go pricing, IaaS vs PaaS vs SaaS, and the shared responsibility model, with a bill calculator.

Cloud computing means renting computing resources (servers, storage, databases, networks, even finished software) from a provider over the internet, instead of buying and running the hardware yourself. You ask for what you need, you get it in seconds, and you pay only for what you use.

Before the cloud, a startup that wanted a website had to buy servers, find a room with cooling and power, and guess how much traffic would arrive. Guess too low and the site crashed; guess too high and money sat idle. The cloud turns that guess into a dial you can turn up or down at any time. This lesson explains the vocabulary you will meet in every later lesson of this tutorial.

The five ideas behind "the cloud"

What makes it "cloud"?

On demand Create a server in seconds, with no ticket and no waiting.
Over the network Everything is reached through the internet or a private link.
Pooled resources Many customers share the provider's huge data centers.
1 / 2

The most important shift is pay-as-you-go. Instead of a large up-front purchase (capital expense), you receive a bill that grows and shrinks with usage (operating expense). That is great for experiments, and it is also why forgotten resources cost real money. You will learn how to control that in Account, Free Tier and Billing Setup.

Three service models: IaaS, PaaS, SaaS

Cloud services differ in how much the provider manages for you. Think of getting dinner:

  • IaaS (Infrastructure as a Service) is buying groceries. You get raw building blocks (virtual machines, disks, networks) and you cook everything yourself. Example on Google Cloud: Compute Engine.
  • PaaS (Platform as a Service) is a meal kit. The platform runs your code and handles servers, patching and scaling; you bring the application. Examples: App Engine, Cloud Run.
  • SaaS (Software as a Service) is a restaurant. You just use finished software. Examples: Gmail and Google Docs in Google Workspace.
flowchart LR
  A["IaaS<br/>You manage OS, runtime, app"] --> B["PaaS<br/>You manage app and data"] --> C["SaaS<br/>You manage your settings"]

The further right you go, the less you manage and the less control you have.

Many teams also talk about serverless: services where you never see a server at all. You deploy a function or a container, the provider starts it when a request arrives and stops it when idle, and the bill follows the requests. Cloud Run and Cloud Run functions are Google Cloud's serverless compute; you will meet them in later chapters.

Public, private and hybrid

A public cloud (Google Cloud, AWS, Azure) is run by a provider for many customers. A private cloud is built and used by one organization in its own data center. A hybrid cloud connects the two, for example a bank that keeps customer records on its own hardware but runs analytics in the public cloud. A multicloud setup uses more than one public provider. This tutorial focuses on the public cloud.

The shared responsibility model

Moving to the cloud does not mean the provider handles all security. Security and operations are split, and the split depends on the service model.

Layer IaaS (VM) PaaS (Cloud Run) SaaS (Workspace)
Physical data center Provider Provider Provider
Hardware and network Provider Provider Provider
Operating system patches You Provider Provider
Runtime and platform You Provider Provider
Your application code You You Provider
Your data and access rules You You You

Two rows never move: your data and who is allowed to access it are always your job. Most real-world cloud breaches come from customer-side mistakes such as a public storage bucket or a leaked password, not from the provider's buildings being broken into. We return to this in Common Mistakes and Troubleshooting.

Why teams choose the cloud

  • Speed: a database or a hundred servers in minutes.
  • Elasticity: capacity follows demand, so a sale day does not need hardware bought for it.
  • Global reach: run close to users on every continent (see Regions, Zones and Latency).
  • Managed services: the provider operates databases, queues and AI models for you.
  • Reliability: data centers designed for failure, with replication built in.

The trade-offs are real too: costs can surprise you, you depend on a provider, and you must learn its tools and security model.

A stand-in: modelling a pay-per-use bill

Nothing here touches a real cloud; the code below is a small model of how metering works. Real providers measure many more things, and real prices vary by region and product, so the prices in this example are invented round numbers chosen only to show the arithmetic.

JS Pay-per-use bill model

Press Run (or Ctrl + Enter) to see the console output here.

No output.

The second month costs more, but only for the two days the extra servers existed. That is elasticity expressed as arithmetic.

JS Who is responsible? (shared responsibility lookup)

Press Run (or Ctrl + Enter) to see the console output here.

No output.

The first stand-in is a model of billing; the second is a lookup table of the table above. Neither calls any real service.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. Which service model gives you a virtual machine where you install and patch the operating system yourself?

    IaaS provides raw infrastructure such as virtual machines, so the operating system and everything above it is yours to manage.
  2. 2. Whose job is it to decide who can read the data you store in a cloud service?

    Your data and access rules stay your responsibility whether you use IaaS, PaaS or SaaS. The provider secures the underlying platform.
  3. 3. What does pay-as-you-go mainly change compared with buying servers?

    Metered billing lets cost grow and shrink with usage. It also means idle resources you forgot still produce a bill.
  4. 4. Gmail and Google Docs are examples of which model?

    They are finished applications you use directly, which is the definition of SaaS.

Summary

  • Cloud computing rents on-demand, elastic, metered resources over a network.
  • IaaS gives building blocks, PaaS runs your code, SaaS gives finished software; serverless hides servers entirely.
  • Public, private, hybrid and multicloud describe where the resources live.
  • Under shared responsibility, the provider secures the platform, but your data and access rules are always yours.
  • Pay-as-you-go is a strength and a risk: unused resources still bill.

Next: Google Cloud at a Glance