AWS Global Infrastructure

The AWS global infrastructure is the physical foundation the cloud runs on, organized into Regions, Availability Zones, and edge locations. On the Cloud Practitioner (CLF-C02) exam, you are rarely asked to memorize how many Regions exist. Instead, you are tested on why the structure exists: how spreading resources across Availability Zones gives you resilience, how edge locations give you low latency, and how you pick the right Region for a workload.

Core Idea

• A Region is an isolated geographic area (like us-east-1 in Northern Virginia) made up of multiple, physically separate Availability Zones.
• An Availability Zone (AZ) is one or more discrete data centers with independent power, cooling, and networking; deploying across multiple AZs is the standard way to achieve high availability.
• Edge locations sit outside Regions and AZs and cache content close to users through Amazon CloudFront for low-latency delivery.

Regions

A Region is a separate geographic area, and each Region is fully isolated from the others. This isolation is deliberate: it supports fault containment and, critically, data residency and compliance — data you place in a Region stays in that Region unless you explicitly move it. Every Region contains at least three Availability Zones (most have three or more), and AZs within a Region are connected by low-latency, high-bandwidth private links. Not every AWS service is available in every Region, and new services often launch in a subset of Regions first.

Availability Zones

An Availability Zone is one or more discrete data centers with redundant power, networking, and connectivity, housed in separate facilities. AZs in the same Region are far enough apart to avoid a single disaster affecting more than one, yet close enough for synchronous replication and low-latency connections. The exam's central lesson: to build a highly available application, deploy your resources across at least two AZs so the failure of a single AZ (or a single data center) does not take your application down. Services like Elastic Load Balancing and Auto Scaling are designed to distribute workloads across multiple AZs automatically.

Edge Locations, CloudFront, Local Zones, and Wavelength

Edge locations are a large, separate set of sites — far more numerous than Regions — used mainly by Amazon CloudFront, the content delivery network (CDN). CloudFront caches copies of content (images, video, static files, and more) at edge locations near end users, cutting latency and reducing load on the origin. Route 53 and AWS Global Accelerator also use the edge network. Local Zones place compute, storage, and select services physically closer to large population or industry centers for single-digit-millisecond latency to nearby users. AWS Wavelength embeds AWS infrastructure inside 5G telecom networks so mobile and edge-device traffic reaches applications without leaving the carrier network.

How to Choose a Region

Four factors drive Region selection on the exam:

• Compliance and data residency — legal or regulatory rules may require data to stay in a specific country or geography. This factor usually wins when present.
• Latency (proximity to users) — choose a Region close to your end users to reduce response times.
• Service availability — confirm the services and features you need are offered in that Region.
• Pricing — costs vary by Region, so price can be a tiebreaker.

High Availability vs. Fault Tolerance vs. Disaster Recovery

These three terms sound alike but are tested as distinct ideas:

• High availability (HA) — the system stays operational with minimal downtime, often by using multiple AZs so a failure is quickly recovered from.
• Fault tolerance — the system continues operating with no interruption even when a component fails, through full redundancy of components.
• Disaster recovery (DR) — the plan and process for restoring service after a major outage or disaster, measured by RTO (recovery time objective) and RPO (recovery point objective).

High-Yield Exam Patterns

• "Achieve high availability" or "survive a data center failure" → deploy across multiple Availability Zones, not multiple Regions.
• "Reduce latency for global users" or "cache static content" → CloudFront and edge locations.
• "Data must stay in a specific country" → this is compliance / data residency, and it decides the Region.
• "One or more discrete data centers" is the exam's definition of an Availability Zone.
• "Bring AWS into a 5G network" → AWS Wavelength; "AWS closer to a specific city for low latency" → Local Zone.
• Regions are isolated from each other; resources are not replicated across Regions automatically.

Common Traps to Avoid

• Confusing an AZ with a single data center — an AZ can be one or more data centers, and an edge location is neither an AZ nor a Region.
• Choosing multiple Regions for basic high availability when multiple AZs are the intended, cheaper answer.
• Assuming a service is available in every Region — service availability varies by Region.
• Mixing up fault tolerance (zero interruption) with high availability (minimal downtime, quick recovery).
• Thinking edge locations run full workloads — they primarily cache and deliver content, they are not general compute.