AWS Global Infrastructure — Study Notes

AWS Academy · Module 1

AWS Global
Infrastructure

A secure, extensive, and reliable platform offering more than 200 fully featured services from data centers around the world.

32 Regions & 102 AZs 410+ Points of Presence Single-Digit MS Latency

AWS Global Infrastructure Topics Title slide

01 — Infrastructure Topics

AWS Global Cloud Infrastructure

The AWS Global Cloud Infrastructure enables highly available, scalable, and flexible architectures, allowing you to deploy workloads close to your customers.

AWS Infrastructure Topics Overview Slide

Slide 1 · AWS Infrastructure Topics — Regions, AZs, Local Zones, Data Centers, and PoPs

Main Components

🌍

AWS Regions

Physical geographic locations consisting of two or more Availability Zones, isolated for fault tolerance.

🏢

Availability Zones (AZs)

Isolated locations within a Region containing one or more data centers, designed for high availability and resilience.

📍

AWS Local Zones

Extension of AWS infrastructure closer to end users to run latency-sensitive workloads in specific geographies.

🛡️

AWS Data Centers

Highly secure physical facilities hosting tens of thousands of active online servers and network hardware.

⚡

AWS Points of Presence (PoPs)

Over 410 edge locations and regional edge caches used for global caching and Route 53 DNS delivery.

Key Facts & Considerations

AWS currently spans 102 Availability Zones across 32 geographic Regions.
Allows you to deploy workloads close to customers for lower latency and compliance.
Choose the right Region based on business requirements, compliance, and workload needs.
Use Points of Presence (PoPs) and edge caches when single-digit millisecond latency is critical.

💡

Best Practice

Always select infrastructure locations based on your specific latency, compliance, and availability requirements.

02 — Geographic Regions

Selecting Regions

Understand how AWS Regions function and the key criteria to choose the best Region for your data and workloads.

Slide 2 · Selecting Regions — Isolated geographic areas containing multiple Availability Zones

🌍

Region Characteristics

A geographical area composed of two or more AZs.
Each AZ consists of one or more data centers.
Regions are completely isolated from each other for fault tolerance.
Connected via AWS private global backbone network (lower cost, more consistent latency than public internet).
You control and enable data replication across Regions (it is not automatic).

🔒

Availability & Access

Most Regions are enabled by default.
Newer Regions (introduced after March 20, 2019) are disabled by default — you must manually enable them.
Some Regions have restricted access (e.g., AWS GovCloud (US) for government workloads and compliance needs).

❓

Key Considerations

Compliance and data sovereignty requirements.
Network latency to your users or customers.
Service availability (not all services are available in every Region).
Business continuity and disaster recovery needs.

📌

Best Practice & Responsibility

Always choose your Region based on latency, compliance, and regulatory requirements. **You are responsible for deciding where your data resides.**

03 — Availability Zones

Selecting Availability Zones (AZs)

Learn how AZs provide physical isolation and redundancy to build highly resilient, fault-tolerant cloud architectures.

Slide 3 · Selecting Availability Zones — Isolated locations physically separated within a metropolitan area

🏢

AZ Characteristics

Designed for fault isolation (independent failure zones).
Physically separated within a metropolitan area.
Located in lower-risk floodplains.
Have independent power supplies and backup generation.
Fed by different utility grids to avoid single points of failure.
Interconnected with other AZs in the same Region using high-speed private links.

⚙️

Best Practices

For certain services (like Amazon EC2), you can choose specific Availability Zones during deployment.
AWS strongly recommends replicating across multiple Availability Zones for high resiliency.
Design your applications to survive the temporary or prolonged failure of a single Availability Zone.

🎁

Benefits

Distributing workloads protects against natural disasters, power failures, or system issues.
Improves overall availability and fault tolerance.
No data center belongs to more than one Availability Zone, preventing shared physical failures.

04 — Edge Scaling

Using AWS Local Zones

Run latency-sensitive portions of your workloads closer to end users to deliver single-digit millisecond latency applications.

Slide 4 · Using Local Zones — Extension of an AWS Region closer to large population centers

📍

Key Features

Extension of an AWS Region closer to population centers.
Place compute, storage, databases, and services where no full Region exists.
Enable single-digit millisecond latency applications.
Managed and fully supported by AWS.

🕹️

Common Use Cases

Media & entertainment content creation
Real-time multiplayer gaming
Reservoir simulations
Electronic design automation
Machine Learning (ML) inference at the edge

⚡

How They Work & Benefits

Provide high-bandwidth, secure connection to parent Region.
Seamlessly integrate using the same VPC, EC2, EBS, and ELB APIs.
Pay only for resources used.
Consistent AWS deployment experience.

05 — Physical Layer

Role of AWS Data Centers

Data centers are the physical foundation of the AWS Global Infrastructure — hosting physical servers, switches, and storage equipment.

Slide 5 · Role of AWS Data Centers — Online, active physical facilities with custom ODM equipment

🛡️

Key Characteristics

Each data center typically contains tens of thousands of servers.
All data centers are active and serving customer traffic (no idle backup sites).
You do not choose specific data centers when deploying; you choose Regions and AZs instead.

🛠️

Redundant Design

Built with redundancy (N+1 configuration).
Automated systems reroute traffic away from failures.
Uses custom ODM equipment and customized protocols.
Sufficient spare capacity to balance loads.

⚠️

Important Resiliency Rule

Hosting resources in a single physical location increases risk. Failures, though rare, can affect availability in that data center. Always distribute workloads across multiple Availability Zones for resiliency.

06 — Content Delivery

AWS Points of Presence (PoPs)

Deliver content and domain requests globally with the lowest possible latency using edge locations and regional edge caches.

Slide 6 · AWS Points of Presence — Global edge locations and regional caches

Main Components

Edge Locations

AWS data centers and servers located close to customers. Serve as initial destinations for Amazon CloudFront requests, caching popular content for fast delivery.

Regional Edge Caches

Mid-tier caches located between origin servers and edge locations. Store less frequently accessed content that is not popular enough to stay in edge caches, reducing latency.

Key Facts & Operation

CloudFront uses a global network of over 410 PoPs (400+ edge locations and 13 regional caches).
Present in North America, Europe, Asia, Australia, South America, Middle East, Africa, and China.
Supports services like Amazon CloudFront, Amazon Route 53, and AWS Global Accelerator.
First request goes to the nearest edge location; if not cached, it is fetched from the Regional Edge Cache or origin server.

🎁

Optimized Delivery

Users get faster content delivery while AWS optimizes bandwidth and cost behind the scenes.

AWS Global Infrastructure Key Takeaways Slide

Module 1 · Takeaways

Key Takeaways:
AWS Global Infrastructure

✓
The AWS Global Infrastructure spans Regions, Availability Zones, Local Zones, Data Centers, and Points of Presence.
✓
Regions connect over private backbone networks, and Availability Zones are isolated, redundant data center clusters.
✓
Local Zones bring compute, storage, and databases closer to end-users to enable low-latency applications.
✓
Data Centers are the physical foundation designed with **N+1 redundancy**, while PoPs cache content globally using **edge locations and regional edge caches**.

AWS GlobalInfrastructure

AWS Global Cloud Infrastructure

Main Components

AWS Regions

Availability Zones (AZs)

AWS Local Zones

AWS Data Centers

AWS Points of Presence (PoPs)

Key Facts & Considerations

Best Practice

Selecting Regions

Region Characteristics

Availability & Access

Key Considerations

Best Practice & Responsibility

Selecting Availability Zones (AZs)

AZ Characteristics

Best Practices

Benefits

Using AWS Local Zones

Key Features

Common Use Cases

How They Work & Benefits

Role of AWS Data Centers

Key Characteristics

Redundant Design

Important Resiliency Rule

AWS Points of Presence (PoPs)

Main Components

Edge Locations

Regional Edge Caches

Key Facts & Operation

Optimized Delivery

Key Takeaways:AWS Global Infrastructure

AWS Global
Infrastructure

Key Takeaways:
AWS Global Infrastructure