Skip to content
Tutorials Center
My learning

Topics

Lesson 1 of 70 Β· Getting started with AWS

What is AWS?

Learn what Amazon Web Services is, how IaaS, PaaS and SaaS differ, what Regions, Availability Zones and edge locations are, and how this tutorial teaches AWS without an account.

Amazon Web Services (AWS) is a collection of more than two hundred services that let you rent computing power, storage, databases, networking and many other building blocks over the internet, and pay only for what you use. Instead of buying a server, waiting weeks for it to arrive, and keeping it running in a closet, you ask AWS for one with an API call, and you can have it in under a minute. When you are done, you delete it and the bill stops.

This tutorial takes you from "what is a Region?" to designing highly available systems, writing IAM policies, deploying serverless APIs and keeping the monthly bill under control. This first lesson draws the map: what cloud computing means, how AWS is laid out around the world, and how the services fit together. It also explains how this site teaches AWS, because we cannot give you an AWS account inside a web page.

What cloud computing really changes

Before the cloud, running software meant buying hardware for your peak load, installing it, and paying for it whether it was busy or idle. The cloud replaces that with a few ideas that you will meet again and again:

  • On-demand self-service. You create resources yourself, through a console, a command line or code, with no purchase order.
  • Pay as you go. Most services bill per second, per request or per gigabyte-month. An idle test server costs almost nothing once you turn it off.
  • Elasticity. You can add capacity when traffic grows and remove it when it falls. Doing this automatically is the subject of Auto Scaling Groups.
  • Global reach. You can run the same application in several parts of the world with the same tools.
  • Managed services. For many jobs (databases, queues, search) AWS runs the machines, patches them and replaces failed parts, so you only work with the service itself.

The trade-off is that the bill is now variable and the responsibility for configuration is yours. A forgotten resource keeps costing money, and a wrongly configured permission can expose data. Much of this tutorial is about doing these two things well.

IaaS, PaaS, FaaS and SaaS

Cloud services are often grouped by how much of the stack you manage yourself.

  • IaaS (Infrastructure as a Service): you rent virtual machines, disks and networks and manage everything above the hypervisor. Amazon EC2 is the classic example (see EC2 Instances).
  • PaaS (Platform as a Service): you hand over your application code and the platform handles servers, scaling and patching. AWS Elastic Beanstalk and AWS App Runner work this way.
  • FaaS (Function as a Service): you upload a function and AWS runs it when an event arrives. AWS Lambda is the example, covered in Lambda Functions.
  • SaaS (Software as a Service): you simply use finished software. Amazon Connect, a ready-made cloud contact center, is an example. For this tutorial, SaaS is background knowledge, since you will mostly build with the other three.

These are labels on a spectrum, not strict boxes. Many AWS services sit between two of them. Still, the question "which layers do I still own?" is the most useful one to ask about any service, and it leads directly to the shared responsibility model.

The stand-in below lists the layers of a stack and shows which ones you manage in each model. It is a simplification (for example, in Lambda you also configure memory, permissions and triggers), but the pattern is accurate: the more managed the service, the fewer layers you operate.

Python Which layers do you manage?

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

Program input Β· what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

How AWS is laid out: Regions, Availability Zones, edge locations

AWS is not one giant data center. It is built from three kinds of locations.

A Region is a geographic area, such as us-east-1 (N. Virginia), eu-west-1 (Ireland) or ap-southeast-2 (Sydney). Each Region is independent: it has its own copy of the services, its own prices, and its own resources. A server you start in Ireland does not exist in Sydney. You choose a Region based on where your users are, the laws about where data may live, which services the Region offers and what it costs.

An Availability Zone (AZ) is one or more separate data centers inside a Region, with their own power, cooling and networking, connected to the other AZs of the same Region by fast, low-latency links. Most Regions have at least three AZs. They are far enough apart that a fire or a power failure usually affects only one, and close enough that your application can write to two of them at once. AZ names look like eu-west-1a, but the letter-to-building mapping is different for each account, so two accounts that both say eu-west-1a may not mean the same building. AZ IDs such as euw1-az1 are the same for everyone, and they matter when you coordinate between accounts.

Edge locations are small sites in many cities that serve cached content close to users. CloudFront, the content delivery network, and the DNS service in Route 53 use them. They are not places where you normally run your servers.

flowchart TB
    W["AWS Global Infrastructure"] --> R1["Region eu-west-1"]
    W --> R2["Region us-east-1"]
    R1 --> A1["AZ 1"]
    R1 --> A2["AZ 2"]
    R1 --> A3["AZ 3"]
    W --> E["Edge locations<br/>cached content near users"]

A Region contains several Availability Zones; edge locations sit outside the Regions and serve cached content.

The practical rule: one AZ is a single point of failure, two or more AZs give you high availability, and two Regions give you disaster recovery. High Availability and Disaster Recovery explores this in depth. The model below shows why spreading across AZs pays off.

Python Spread across Availability Zones

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

Program input Β· what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The service map

AWS groups its services into families. You do not need to memorize them now; this tutorial teaches the important ones in order.

Family Typical services Covered in
Identity and security IAM, KMS, GuardDuty IAM Users, Groups and Roles
Compute EC2, Lambda, ECS, Fargate EC2 Instances
Networking VPC, Route 53, CloudFront, ELB VPC Fundamentals
Storage S3, EBS, EFS S3 Buckets and Objects
Databases RDS, Aurora, DynamoDB, ElastiCache DynamoDB Fundamentals
Integration SQS, SNS, EventBridge, Step Functions SQS, SNS and Fan-Out
Operations and cost CloudWatch, CloudTrail, Cost Explorer Cost Management and Optimization
Infrastructure as code CloudFormation, CDK CloudFormation

A few services are global (IAM, Route 53, CloudFront), meaning you configure them once for the whole account. Most services are regional, meaning every Region has its own separate set of resources. Forgetting which Region the console is showing is the most common reason a beginner thinks a resource has "disappeared".

How this tutorial works without an AWS account

A web page cannot create real AWS resources. So every lesson is honest about what is real and what is not:

  • Real AWS material is read-only. Console steps, aws CLI commands, SDK code and JSON or YAML configuration appear in ordinary code blocks. They are accurate and you can copy them into your own account, but the Run button is not offered on them.
  • Runnable stand-ins show the idea. The python examples you can edit and run are small emulators written for this site: an S3-like object store, an IAM policy evaluator, a pricing calculator, a queue with a visibility timeout and more. They behave like the real thing in the aspects the lesson explains, and they are not AWS code. Each lesson says which parts are real and which are the stand-in.

If you want to try the real thing, the next lesson shows how to open and secure an account, and the free tier and credits help keep early experiments cheap.

Check your understanding

Quick check Β· 4 questions

of 4 answered You scored / 4 Β· perfect!

  1. 1. Which statement about an Availability Zone is correct?

    An Availability Zone is a set of isolated data centers within a Region. Regions are geographic areas that contain several AZs, and edge locations are the cache sites.
  2. 2. You rent virtual machines and install your own operating system updates. Which model is this?

    With IaaS (for example EC2) AWS runs the hardware and virtualization, and you manage the operating system and everything above it.
  3. 3. A server you started in eu-west-1 does not show up in the console. What is the most likely reason?

    Most services are regional, so each Region holds its own resources. Check the Region selector in the console.
  4. 4. How do you get high availability for a web application inside one Region?

    Spreading the application over several AZs means one AZ failure leaves the rest running. Edge locations only cache content.

Summary

  • AWS rents computing, storage, databases and networking on demand, billed by usage.
  • IaaS, PaaS, FaaS and SaaS describe how many layers of the stack you still manage.
  • A Region is an independent geographic area; an Availability Zone is a set of isolated data centers inside it; edge locations cache content near users.
  • Most services are regional; IAM, Route 53 and CloudFront are global.
  • On this site, real AWS commands and code are read-only, and runnable Python emulators illustrate how the services work.

Next: Creating and Securing an Account.