AWS Global Infrastructure – Regions, Availability Zones, Edge, Outposts, Local Zones, and Wavelength Zones

Purpose of This Lesson

This lesson explains the major building blocks of the AWS global infrastructure: Regions, Availability Zones, subnets, edge locations, regional edge caches, Local Zones, Wavelength Zones, and AWS Outposts.

For the AWS Certified Solutions Architect – Professional SAP-C02 exam, this topic matters because many architecture questions are built around latency, high availability, disaster recovery, data residency, hybrid infrastructure, and service placement. You do not need to memorize the exact number of AWS Regions, Availability Zones, or edge locations because those numbers change frequently. You do need to understand what each component is used for and how to choose the right architecture pattern for a given scenario.

Key Concepts

AWS Regions

An AWS Region is a separate geographic area where AWS operates cloud infrastructure. Examples include regions in North America, Europe, Asia Pacific, South America, the Middle East, and other parts of the world.

Regions are physically separate from each other and designed to operate independently. This independence is important for resiliency, compliance, and disaster recovery.

From an architecture perspective, a Region is usually the first major placement decision you make.

Common reasons to choose one Region over another include:

• Proximity to users
• Data residency or compliance requirements
• Service availability
• Cost differences
• Disaster recovery design
• Integration with existing on-premises environments
• Latency-sensitive workloads

For the SAP-C02 exam, remember that AWS Regions are isolated from each other. Deploying across multiple Regions can improve disaster recovery, but it also adds cost, complexity, data replication requirements, and operational overhead.

Availability Zones

An Availability Zone, or AZ, is one or more discrete data centers within an AWS Region. Each Region contains at least two Availability Zones.

Availability Zones are physically separated from each other within a Region but connected using low-latency, high-throughput, redundant networking. This allows applications to run across multiple AZs while still behaving like they are in the same regional environment.

Each Availability Zone has redundant infrastructure, including:

• Power
• Networking
• Cooling
• Physical data center facilities

The main purpose of Availability Zones is high availability. If one AZ has an issue, workloads running in another AZ can continue operating if the architecture was designed correctly.

Common multi-AZ patterns include:

• EC2 instances spread across multiple AZs behind an Elastic Load Balancer
• Amazon RDS Multi-AZ deployments
• Auto Scaling groups spanning multiple subnets in different AZs
• NAT Gateways deployed per AZ for resilient outbound internet access
• Multi-AZ Amazon EKS or ECS worker node placement
• Application tiers split across public and private subnets in multiple AZs

Subnets Are Availability Zone-Scoped

A subnet exists inside a single Availability Zone. This is an important AWS networking concept.

A VPC is regional, but subnets are AZ-specific. When you create subnets, you select the Availability Zone where each subnet will live.

For example, a highly available VPC design might include:

• Public subnet in AZ A
• Private application subnet in AZ A
• Private database subnet in AZ A
• Public subnet in AZ B
• Private application subnet in AZ B
• Private database subnet in AZ B

This lets you deploy resources across multiple AZs while keeping the network layout consistent.

A subnet does not span multiple Availability Zones. If an exam answer suggests that a single subnet provides multi-AZ resiliency, that answer is misleading.

Public and Private Subnets

Inside an Availability Zone, you can create public and private subnets.

A public subnet usually has a route to an internet gateway and is used for resources that must be directly reachable from the internet, such as load balancers or bastion hosts.

A private subnet does not have a direct inbound route from the internet. It is commonly used for application servers, databases, internal services, and workloads that should not be publicly exposed.

In a professional-level architecture, public subnets are usually used sparingly. A common design is to place internet-facing load balancers in public subnets and keep application and database resources in private subnets.

AWS Global Network Backbone

AWS Regions are connected through the AWS global network backbone. This is a high-bandwidth, redundant private network used by AWS to connect its global infrastructure.

This matters for architectures that involve:

• Multi-Region disaster recovery
• Cross-Region replication
• Global application delivery
• Hybrid connectivity
• CloudFront origin access
• Inter-Region service communication

The AWS global network backbone helps provide reliable connectivity between AWS locations, but architects still need to account for latency, data transfer costs, replication design, and failure domains.

AWS Outposts

AWS Outposts extends AWS infrastructure into an on-premises data center or colocation facility. With Outposts, AWS provides dedicated hardware that runs AWS services locally while still being managed as part of the AWS ecosystem.

Use AWS Outposts when an organization needs AWS services on-premises because of:

• Ultra-low latency to local systems
• Local data processing requirements
• Data residency or regulatory constraints
• Migration strategies where workloads cannot immediately move fully to the cloud
• Applications that depend on nearby on-premises systems
• Consistent AWS APIs and tooling across cloud and on-premises environments

Outposts is not simply “another Region.” It is an extension of an AWS Region into a customer-controlled location. It depends on connectivity back to the parent Region for management and service integration.

AWS Local Zones

AWS Local Zones place selected AWS services closer to large population centers, metro areas, or specific user locations. They are designed to reduce latency when the main AWS Region is too far away from the end users or workloads.

Local Zones are useful for workloads that require single-digit millisecond latency to users but do not necessarily need to run inside the customer’s own data center.

Common Local Zone use cases include:

• Media and entertainment workloads
• Real-time gaming
• Interactive streaming
• Virtual workstations
• Low-latency application access
• Latency-sensitive machine learning inference
• Applications serving users in a specific metro area

A Local Zone is associated with a parent AWS Region. You still manage resources through the AWS Region, but selected infrastructure runs closer to the end users.

AWS Wavelength Zones

AWS Wavelength Zones are designed for ultra-low-latency applications accessed over 5G networks. Wavelength places AWS compute and storage services at the edge of telecommunications providers’ 5G networks.

The main use case is reducing latency between mobile devices and application workloads.

Common AWS Wavelength use cases include:

• Augmented reality
• Virtual reality
• Live video processing
• Connected vehicles
• Smart factories
• IoT applications
• Real-time machine learning inference
• Mobile gaming
• Applications requiring single-digit millisecond latency over 5G

The key exam distinction is that Wavelength is specifically tied to 5G network access and telecom edge locations.

Edge Locations

Edge locations are part of the Amazon CloudFront content delivery network. They are used to cache and deliver content closer to users around the world.

When a user requests content through CloudFront, AWS can serve cached content from a nearby edge location instead of sending every request back to the origin.

Common content types served through CloudFront include:

• Images
• Videos
• Static website assets
• JavaScript files
• CSS files
• API responses
• Downloadable files

Edge locations improve user experience by reducing latency and offloading traffic from the origin.

Regional Edge Caches

Regional edge caches sit between CloudFront edge locations and the origin. They provide a larger caching layer that helps improve cache hit ratios and reduce origin load.

A simplified flow looks like this:

1. User requests content.
2. CloudFront checks the nearest edge location.
3. If the content is not cached there, CloudFront may check a regional edge cache.
4. If the content is still not available, the request goes back to the origin.

Regional edge caches are especially useful for content that may not be popular enough to remain cached at every edge location but is still requested often enough to benefit from caching closer to users.

Amazon CloudFront as a Global Delivery Service

Amazon CloudFront is AWS’s content delivery network service. It uses edge locations and regional edge caches to deliver content with lower latency.

CloudFront is a better fit than deploying duplicate application stacks around the world when the main requirement is fast delivery of cacheable content.

Use CloudFront for:

• Static website acceleration
• Video delivery
• Global content distribution
• Reducing origin load
• Improving performance for geographically distributed users
• Adding edge-level security controls
• Fronting applications with AWS WAF

CloudFront is not the same as deploying workloads into multiple Regions, Local Zones, or Wavelength Zones. It primarily accelerates content delivery and request routing at the edge.

Exam-Relevant Takeaways

For the SAP-C02 exam, focus less on memorizing infrastructure counts and more on service selection.

Know these distinctions:

• Region: A geographically separate AWS location used for workload placement, compliance, latency, and disaster recovery.
• Availability Zone: One or more data centers inside a Region, used for high availability and fault isolation.
• Subnet: A network segment inside one Availability Zone.
• AWS Outposts: Runs AWS services on dedicated hardware in an on-premises location.
• AWS Local Zones: Extends selected AWS services closer to users in specific metro areas for low latency.
• AWS Wavelength: Places AWS services at the 5G telecom edge for ultra-low-latency mobile applications.
• CloudFront edge locations: Cache and deliver content closer to users.
• Regional edge caches: Larger CloudFront caching layer between edge locations and origins.

For high availability inside a Region, use multiple Availability Zones.

For disaster recovery across geographically separate locations, use multiple Regions.

For low-latency content delivery, use CloudFront.

For low-latency compute near a metro area, consider Local Zones.

For low-latency mobile/5G workloads, consider Wavelength Zones.

For AWS services inside a customer data center, consider AWS Outposts.

Architecture Decision Guide

Scenario	Best AWS Choice	Why
Need high availability for an application inside one AWS Region	Deploy across multiple Availability Zones	AZs provide fault isolation within a Region and are connected by low-latency networking
Need disaster recovery in a geographically separate location	Multi-Region architecture	Regions are independent and geographically separated
Need public-facing application entry point while keeping servers private	Public subnets for load balancers, private subnets for application servers	Reduces public exposure while still allowing internet access through controlled entry points
Need subnets for workloads in two AZs	Create separate subnets in each AZ	A subnet is scoped to one Availability Zone and cannot span AZs
Need AWS services physically located in an on-premises data center	AWS Outposts	Provides AWS-managed hardware and AWS APIs in the customer location
Need very low latency to users in a specific metro area	AWS Local Zones	Extends selected AWS services closer to end users
Need ultra-low latency for mobile devices over 5G	AWS Wavelength	Places AWS services at the telecom 5G edge
Need faster global delivery of images, videos, and static files	Amazon CloudFront	Caches content at edge locations closer to users
Need to reduce origin load for global content distribution	CloudFront with regional edge caches	Regional edge caches improve cache efficiency before requests return to the origin
Need consistent AWS-style operations on-premises during a hybrid migration	AWS Outposts	Allows use of AWS APIs and services while keeping workloads local

Common Exam Traps

Trap 1: Thinking a Subnet Spans Multiple Availability Zones

A VPC is regional, but a subnet belongs to one Availability Zone. To build a multi-AZ design, create multiple subnets across multiple AZs.

Trap 2: Confusing Multi-AZ with Multi-Region

Multi-AZ improves availability within a Region. Multi-Region improves geographic resiliency and disaster recovery.

If the question is about surviving an AZ failure, use a multi-AZ design.

If the question is about surviving a regional outage or meeting geographic DR requirements, use multiple Regions.

Trap 3: Choosing Outposts for General Global Latency

AWS Outposts is for running AWS infrastructure in an on-premises location. It is not the default answer for delivering content globally or reducing latency to internet users.

For global content delivery, think CloudFront.

For metro-area low-latency compute, think Local Zones.

For 5G mobile edge workloads, think Wavelength.

Trap 4: Confusing Local Zones and Wavelength Zones

Local Zones bring selected AWS services closer to end users in specific geographic areas.

Wavelength Zones are specifically designed for applications accessed over 5G networks through telecom providers.

If the exam scenario mentions mobile devices, 5G, telecom edge, AR/VR, or connected vehicles, Wavelength is likely relevant.

Trap 5: Assuming Exact Infrastructure Counts Matter

AWS constantly adds Regions, AZs, edge locations, Local Zones, and other infrastructure. The SAP-C02 exam does not require memorizing exact global infrastructure counts.

Understand the architecture role of each component instead.

Trap 6: Using CloudFront When Compute Must Run Near the User

CloudFront is excellent for caching and delivering content. However, if application compute must run close to users, CloudFront alone may not be enough.

For low-latency compute placement, evaluate Local Zones, Wavelength, Outposts, or multi-Region deployment depending on the scenario.

Trap 7: Forgetting That Not Every Service Is Available Everywhere

Not all AWS services are available in every Region, Local Zone, or Wavelength Zone. In real designs, always verify service availability before committing to a location strategy.

Real-World Engineer Notes

In production architecture, global infrastructure decisions affect cost, resiliency, compliance, troubleshooting, and operations.

Region Selection Is a Business and Technical Decision

Choosing a Region is not just about picking the closest location. You also need to consider compliance, service availability, disaster recovery, pricing, and integration with existing systems.

For example, a workload may serve users in one country but need to store data in another jurisdiction due to legal requirements. In that case, the Region decision becomes a governance and compliance issue, not just a latency issue.

Multi-AZ Should Be a Default for Critical Workloads

For production workloads, single-AZ designs are usually a risk unless there is a strong cost or simplicity reason. A professional-level AWS design normally spreads critical resources across multiple AZs.

Examples include:

• Load balancer across at least two public subnets
• Application Auto Scaling group across at least two private subnets
• RDS Multi-AZ for relational database availability
• NAT Gateway per AZ for resilient outbound connectivity
• Proper route tables per AZ to avoid cross-AZ dependency where possible

Multi-Region Adds Complexity

Multi-Region architecture can improve resiliency, but it creates new challenges:

• Data replication
• DNS failover
• Application consistency
• Database conflict handling
• Cross-Region costs
• Deployment automation
• Monitoring across Regions
• Security policy consistency
• Backup and recovery testing

For the exam, do not automatically choose multi-Region unless the scenario requires regional disaster recovery, global user access, or strict geographic separation.

Edge Services Reduce Latency Without Full Regional Duplication

CloudFront can significantly improve user experience without deploying full infrastructure stacks around the world. This is especially useful for static assets, downloads, media, and cacheable application responses.

A common real-world pattern is:

• Application runs in one or more Regions.
• CloudFront sits in front of the application.
• AWS WAF protects the edge.
• Static content is cached globally.
• Dynamic requests are routed back to the origin.

This provides global performance benefits while keeping the backend architecture simpler.

Outposts Requires Physical Planning

AWS Outposts is not just a cloud setting. It involves physical hardware, power, cooling, rack space, network connectivity, and operational planning.

In a real environment, Outposts may be useful when an application must remain on-premises but the organization wants AWS-like management, APIs, and operational consistency.

However, it is not as simple or elastic as launching resources in a normal AWS Region.

Local Zones and Wavelength Are Specialized

Local Zones and Wavelength Zones are powerful, but they are not default choices for ordinary workloads.

Use them when the scenario specifically requires very low latency and the user or device location makes a normal AWS Region too far away.

Wavelength is especially specialized because it depends on telecom 5G integration.

Quick Reference Summary

• AWS global infrastructure is made up of Regions, Availability Zones, edge locations, regional edge caches, Local Zones, Wavelength Zones, and Outposts.
• A Region is a geographically separate AWS location.
• An Availability Zone is one or more data centers inside a Region.
• A VPC is regional.
• A subnet is tied to one Availability Zone.
• Use multiple AZs for high availability within a Region.
• Use multiple Regions for disaster recovery across geographic areas.
• Use CloudFront edge locations to cache and deliver content closer to users.
• Use regional edge caches to improve CloudFront cache efficiency and reduce origin requests.
• Use AWS Outposts to run AWS services on-premises.
• Use Local Zones for low-latency workloads near end users in specific metro areas.
• Use Wavelength Zones for ultra-low-latency 5G mobile edge workloads.
• Do not memorize exact counts of AWS Regions, AZs, or edge locations; they change over time.

Flashcards

Q: What is an AWS Region?
A: A geographically separate AWS location that contains multiple Availability Zones and operates independently from other Regions.

Q: What is an Availability Zone?
A: One or more discrete data centers inside an AWS Region, designed for fault isolation and connected to other AZs in the Region with low-latency networking.

Q: Does a subnet span multiple Availability Zones?
A: No. A subnet exists in exactly one Availability Zone.

Q: What AWS infrastructure component should you use for high availability within a Region?
A: Multiple Availability Zones.

Q: What AWS infrastructure pattern should you use for disaster recovery across geographically separate locations?
A: Multi-Region architecture.

Q: What is AWS Outposts used for?
A: Running AWS services on dedicated AWS-managed hardware in an on-premises data center or colocation facility.

Q: What is the purpose of AWS Local Zones?
A: To place selected AWS services closer to end users in specific metro areas for low-latency access.

Q: What is the purpose of AWS Wavelength?
A: To provide ultra-low-latency access to AWS services over 5G telecom networks.

Q: What service uses edge locations and regional edge caches?
A: Amazon CloudFront.

Q: What is the main purpose of CloudFront edge locations?
A: To cache and deliver content closer to users, reducing latency and origin load.

Q: What is a regional edge cache?
A: A larger CloudFront caching layer between edge locations and the origin that improves cache hit rates and reduces origin requests.

Q: Should you memorize the exact number of AWS Regions and edge locations for SAP-C02?
A: No. The numbers change frequently. Understand the purpose and use cases of each infrastructure component.

Q: Which is more appropriate for 5G mobile edge workloads: Local Zones or Wavelength Zones?
A: Wavelength Zones.

Q: Which is more appropriate for caching video files close to users: CloudFront or Outposts?
A: CloudFront.

Q: Which is more appropriate when AWS services must run inside a company data center: Local Zones or Outposts?
A: AWS Outposts.

Practice Questions

Question 1:
A company is deploying a web application in AWS. The application must remain available if a single data center facility or Availability Zone has an outage. Which design should the solutions architect recommend?

A. Deploy all resources into a single subnet in one Availability Zone
B. Deploy resources across multiple Availability Zones in the same Region
C. Deploy the application only to an AWS edge location
D. Deploy the application to AWS Outposts only

Correct Answer:
B. Deploy resources across multiple Availability Zones in the same Region

Explanation:
Availability Zones are designed for fault isolation within a Region. Deploying across multiple AZs improves availability if one AZ has an issue. A single subnet is tied to one AZ and does not provide multi-AZ resiliency.

---

Question 2:
A company wants to deliver images and video files to users around the world with lower latency and reduced load on the origin server. Which AWS service should be used?

A. Amazon CloudFront
B. AWS Outposts
C. AWS Wavelength
D. AWS Direct Connect

Correct Answer:
A. Amazon CloudFront

Explanation:
Amazon CloudFront is AWS’s content delivery network. It uses edge locations and regional edge caches to cache content closer to users and reduce requests to the origin.

---

Question 3:
A manufacturing company needs to run AWS services in its own data center because an application requires very low-latency access to local equipment. The company also wants to use AWS APIs and management tools. Which option is the best fit?

A. AWS Local Zones
B. AWS Outposts
C. Amazon CloudFront
D. AWS Wavelength

Correct Answer:
B. AWS Outposts

Explanation:
AWS Outposts extends AWS infrastructure into an on-premises data center. It is appropriate when workloads need to run locally while still using AWS services, APIs, and operational models.

---

Question 4:
A mobile application requires single-digit millisecond latency to users connected through 5G networks. The workload processes real-time video and augmented reality data. Which AWS infrastructure option is most appropriate?

A. AWS Wavelength
B. AWS Outposts
C. Regional edge caches
D. A single Availability Zone deployment

Correct Answer:
A. AWS Wavelength

Explanation:
AWS Wavelength is designed for ultra-low-latency applications accessed over 5G networks. It places AWS services at the telecom edge, close to mobile users.

---

Question 5:
A solutions architect is designing a VPC across two Availability Zones. Which statement is correct?

A. A subnet can span both Availability Zones
B. A VPC is scoped to a single Availability Zone
C. Each subnet must be created in one specific Availability Zone
D. Edge locations are required before creating private subnets

Correct Answer:
C. Each subnet must be created in one specific Availability Zone

Explanation:
A VPC is regional, but each subnet is associated with one Availability Zone. To design across multiple AZs, create separate subnets in each AZ.