AWS Global Infrastructure
AWS operates a global network of 30+ Regions, 2–6 Availability Zones per Region, and 400+ Edge Locations to deliver low latency, high availability, and data sovereignty. Region selection involves compliance, proximity, service availability, and pricing. Edge services like CloudFront and Global Accelerator leverage this infrastructure for content delivery and network performance.
1. AWS Global Infrastructure Overview
2. AWS Regions
A Region is a geographic area that contains 2 or more Availability Zones (most have 3–6). Each Region is completely independent — data does not automatically replicate between Regions.
Currently: 30+ Regions across North America, Europe, Asia Pacific, Middle East, Africa, and South America.
How to Choose a Region
| Factor | Consideration | Example |
|---|---|---|
| Data Sovereignty / Compliance | Laws may require data to stay within a country | EU data must stay in an EU Region for GDPR |
| Proximity to Users | Closer Region = lower latency | US users → us-east-1 or us-west-2 |
| Service Availability | Not all services are available in every Region | Some newer services launch in us-east-1 first |
| Pricing | Same service costs vary by Region | us-east-1 (N. Virginia) is often the cheapest |
Important: Data does NOT automatically move between Regions — you explicitly configure cross-region replication if needed.
3. Availability Zones (AZs)
Each AZ is one or more discrete, physically separate data centers within a Region:
Key AZ properties:
- Physically separated by meaningful distance (miles apart) to isolate from local disasters
- Connected by high-bandwidth, low-latency private fiber — not the public internet
- Each has independent power, cooling, networking, and physical security
- AZ letter names (us-east-1a) are randomized per AWS account to ensure load is spread evenly
Multi-AZ architecture = High Availability. Deploying across 2+ AZs eliminates single points of failure. If one AZ loses power, your workload continues in the other AZs.
4. Edge Locations
Edge Locations (Points of Presence) are infrastructure nodes used to deliver content and services closer to end users:
| Feature | Details |
|---|---|
| Count | 400+ locations globally — far more than Regions |
| Used by | Amazon CloudFront (CDN), Route 53 (DNS), Lambda@Edge, AWS Global Accelerator |
| Purpose | Cache content and serve requests from the location nearest the user |
| Not full Regions | Cannot run EC2, RDS, etc. — only CDN and DNS infrastructure |
5. CloudFront vs Global Accelerator
Two services that use the global edge network for different purposes:
| Feature | Amazon CloudFront | AWS Global Accelerator |
|---|---|---|
| Purpose | Content Delivery Network (CDN) | Network performance accelerator |
| Caching | Yes — caches static and dynamic content | No caching |
| Use case | Websites, images, video, API responses | Non-cacheable TCP/UDP apps (gaming, VoIP, IoT) |
| IP addresses | DNS-based (dynamic) | 2 static Anycast IP addresses |
| Failover | Cache-miss routing | Automatic failover to healthy endpoints in seconds |
| Integration | WAF, Shield, S3, ALB | ALB, NLB, EC2, Elastic IP |
6. AWS Local Zones
- Extensions of AWS Regions placed in major metropolitan areas (e.g., Los Angeles, Chicago, Dallas, Boston)
- Run a subset of AWS services (EC2, EBS, ECS, RDS) directly in the metro area
- Use case: Gaming, media production, real-time interactive applications needing single-digit millisecond latency
- Named like:
us-west-2-lax-1a(Los Angeles extension of US West Oregon)
7. AWS Wavelength Zones
- AWS compute and storage embedded inside 5G mobile network providers' data centers (Verizon, Vodafone, KDDI, etc.)
- Traffic from mobile devices never leaves the telecom network to reach AWS compute
- Use case: Ultra-low latency applications for mobile and edge computing — AR/VR, autonomous vehicles, real-time analytics on mobile data
8. AWS Outposts
- AWS-managed rack of hardware that AWS installs in your on-premises data center or colocation facility
- Extends AWS infrastructure, APIs, and tools to your premises — run EC2, EBS, RDS, ECS on-site
- Use cases: Data residency regulations requiring data to never leave on-premises, ultra-low latency to on-premises systems, hybrid workloads
- AWS delivers, installs, operates, and maintains the hardware — you just rack the space and provide power
9. Amazon Route 53
Route 53 is AWS's highly available, scalable managed DNS service:
- Translates human-readable domain names (example.com) into IP addresses
- Performs health checks and routes traffic away from unhealthy endpoints
- Routing policies:
| Policy | Routes Traffic Based On |
|---|---|
| Simple | Single destination; no health check |
| Weighted | Percentage split — useful for A/B testing |
| Latency-based | Lowest latency Region for the user |
| Failover | Active/standby — switch on health check failure |
| Geolocation | User's country or continent |
| Geoproximity | Geographic distance with bias |
| Multi-value | Multiple IPs returned; filters unhealthy ones |
10. CLF-C02 Quick Reference
| Topic | Key Fact |
|---|---|
| Regions | 30+ worldwide; independent; data does NOT auto-replicate between them |
| AZs per Region | Usually 3–6; minimum 2 |
| AZ characteristics | Physically separate; independent power/cooling; low-latency private fiber |
| Edge Locations | 400+ globally; used by CloudFront, Route 53, Global Accelerator |
| Region selection factors | Compliance, proximity to users, service availability, pricing |
| Multi-AZ | Eliminates single points of failure; high availability pattern |
| CloudFront | CDN; caches content at edge locations; integrates with WAF |
| Global Accelerator | Routes non-cacheable traffic via AWS backbone; static Anycast IPs |
| Local Zones | Metro area extensions; single-digit ms latency for local users |
| Wavelength Zones | Inside 5G networks; ultra-low latency for mobile edge apps |
| Outposts | AWS hardware on-premises; run AWS services locally |
| Route 53 | Managed DNS; multiple routing policies including latency and failover |
Practice Questions8
Q1. What is an AWS Region?
Select one answer before revealing.
Q2. What is an AWS Availability Zone (AZ)?
Select one answer before revealing.
Q3. Why should you deploy an application across multiple Availability Zones in the same Region?
Select one answer before revealing.
Q4. Users in Southeast Asia are experiencing high latency accessing your application hosted in us-east-1. Which AWS service can cache your static content at locations closest to those users?
Select one answer before revealing.
Q5. What is an AWS Edge Location and which service primarily uses it?
Select one answer before revealing.
Q6. A government agency needs to ensure their data never leaves a specific country due to data sovereignty laws. Which AWS feature primarily enables this compliance requirement?
Select one answer before revealing.
Q7. What is the difference between AWS Local Zones and AWS Wavelength Zones?
Select one answer before revealing.
Q8. Which AWS service routes user traffic over AWS's private global network backbone instead of the public internet, improving performance and providing static IP addresses?
Select one answer before revealing.