4 AWS Cloud Concepts
Learn how AWS cloud computing works, how Regions and Availability Zones support deployment and resilience, what common AWS services do, and how security responsibilities are shared.
On-demand computing and capacity
provides resources such as servers, storage, databases, and networking on demand, typically with pay-as-you-go pricing. Rather than buying and maintaining all hardware in advance, an organization can provision the capacity it needs and change it as requirements change. This can reduce upfront infrastructure costs, speed up experimentation, and make it easier to adjust capacity; actual costs still depend on which resources are used and for how long.
is the ability of a system to handle increased workload by adding capacity. is the ability to adjust capacity up or down as demand changes. For example, a web application might use Auto Scaling to add EC2 instances during a busy period and remove them afterward.
Regions and Availability Zones
An is a separate geographic area. AWS services and resources are often created in a particular Region, so the choice affects factors such as proximity to users and applicable location requirements. Resources are not automatically replicated between Regions; cross-Region replication or recovery generally requires deliberate configuration.
Each Region contains multiple independent Availability Zones (AZs). An is an isolated location within a Region, made up of one or more data centers. AZs are connected by low-latency networking. Deploying parts of an application in more than one AZ can help it continue operating if an issue affects one AZ, but it does not by itself protect against a disruption affecting the whole Region.
For example, a team could run application servers in two AZs in one Region and use a load balancer to distribute traffic between them. If it also needs a recovery plan for a Region-level disruption, it must design and test a separate multi-Region approach.
Common AWS services
AWS offers services for different parts of an application. Common examples include:
Virtual servers: Amazon EC2 provides resizable compute capacity.
Serverless code execution: AWS Lambda runs code without requiring customers to manage servers.
Object storage: Amazon S3 stores and retrieves objects such as files, backups, and static website assets.
Relational databases: Amazon RDS helps set up and operate relational databases.
Networking: Amazon VPC provides a virtual network where customers can launch and configure AWS resources.
Traffic distribution: Elastic Load Balancing distributes incoming traffic across targets.
Identity and permissions: AWS Identity and Access Management (IAM) controls access to AWS resources.
Monitoring: Amazon CloudWatch collects and presents monitoring data for AWS resources and applications.
Services can be combined. For instance, a website might use a load balancer to route requests to EC2 instances, store uploaded files in S3, and keep application data in RDS. Service capabilities and availability can vary, so check the documentation for the chosen Region and use case.
AWS and the customer share responsibility for security and compliance. AWS describes this division as security “of” the cloud and security “in” the cloud.
AWS: security of the cloud. AWS protects the underlying infrastructure that runs its services, including facilities, hardware, networking, and virtualization.
Customer: security in the cloud. Customers manage security tasks associated with their workloads. Depending on the services used, these tasks may include protecting data, configuring access permissions and network controls, and maintaining operating systems and applications.
The division depends on the service. With a virtual machine such as EC2, the customer has more operating-system and application-management responsibilities. With a more managed service, AWS handles more of the underlying operations, but customers remain responsible for appropriate data handling, access, and configuration. Using AWS does not automatically make an application secure; customers must understand and fulfill their part of the model.