AWS Global Infrastructure and Distributed Computing

This guide explores how AWS provides a globally distributed platform to support high-performing, resilient, and low-latency applications through its physical infrastructure and specialized edge services.

Learning Objectives

After studying this guide, you should be able to:

• Differentiate between Regions, Availability Zones (AZs), and Edge Locations.
• Identify the core AWS services provided at the edge to reduce latency and improve security.
• Explain the importance of decoupling in distributed systems using SQS and SNS.
• Select appropriate scaling strategies (Horizontal vs. Vertical) for distributed workloads.
• Understand how AWS Global Accelerator and Amazon CloudFront optimize global traffic.

Key Terms & Glossary

• Region: A physical location in the world where AWS has multiple Availability Zones. Each region is geographically isolated.
• Availability Zone (AZ): One or more discrete data centers with redundant power, networking, and connectivity in an AWS Region.
• Edge Location: Sites that AWS uses to cache content and reduce latency for end users, often used by CloudFront and Route 53.
• Decoupling: An architectural approach where components of a system remain independent so that the failure or scaling of one does not directly impact the other.
• Microservices: A design pattern where an application is built as a collection of small, independent services.

The "Big Idea"

[!IMPORTANT] Distributed computing on AWS is built on the principle of geographical abstraction. By moving logic and data closer to the user (via Edge Locations) and spreading workloads across isolated failure zones (Regions/AZs), applications achieve "The Cloud Trinity": Low Latency, High Availability, and Infinite Scalability.

Formula / Concept Box

Hierarchical Outline

1. AWS Physical Infrastructure
   • Global Regions: Compliance, data sovereignty, and proximity.
   • Availability Zones: Fault isolation and high availability within a region.
   • Local Zones / Outposts: Extending AWS to specific geographic locations or on-premises.
2. Edge Networking & Performance
   • Amazon CloudFront: Content Delivery Network (CDN) for caching static/dynamic content.
   • AWS Global Accelerator: Uses the AWS private network to optimize the path to your application.
   • Amazon Route 53: Scalable DNS and health checking.
3. Distributed System Patterns
   • Asynchronous Messaging: Using Amazon SQS (queuing) and Amazon SNS (pub/sub).
   • Orchestration: Managing containers with Amazon ECS or EKS.
   • Serverless: Event-driven computing with AWS Lambda and Fargate.

Visual Anchors

AWS Infrastructure Hierarchy

Request Flow: Edge vs. Origin

Definition-Example Pairs

• Service Quotas: The maximum number of resources you can create in an AWS account.
  • Example: An AWS account might have a default limit of 200 running On-Demand EC2 instances per region.
• Read Replicas: A copy of a database used to offload read traffic from the primary instance.
  • Example: A WordPress site experiences high traffic; the admin creates an RDS Read Replica so the database can handle more "view" requests without slowing down "write" requests.
• Loose Coupling: Reducing the dependencies between components so they can scale independently.
  • Example: A web server places a photo-processing job into an SQS Queue instead of sending it directly to a worker server. If the worker server crashes, the job stays in the queue until a new worker is ready.

Worked Examples

Problem: Global Low-Latency Video Streaming

Scenario: A company based in London wants to serve high-definition video to users in Tokyo and New York with minimal buffering.

Step-by-Step Breakdown:

1. Storage: Upload video files to an Amazon S3 bucket in the eu-west-2 (London) region.
2. Distribution: Create an Amazon CloudFront distribution. Set the S3 bucket as the "Origin."
3. Edge Delivery: When a user in Tokyo requests a video, CloudFront routes the request to the Edge Location in Tokyo.
4. Caching: If it's the first request, CloudFront fetches the file from London once and caches it in Tokyo. Subsequent Tokyo users get the file instantly from the local cache.
5. Security: Attach AWS WAF (Web Application Firewall) to the CloudFront distribution at the edge to block malicious SQL injection attacks before they reach the London origin.

Checkpoint Questions

1. Which AWS infrastructure component consists of one or more discrete data centers?
   • Answer: Availability Zone (AZ).
2. If you need to send a message to multiple subscribers simultaneously (Fan-out pattern), which service should you use?
   • Answer: Amazon SNS (Simple Notification Service).
3. What is the primary difference between AWS Global Accelerator and Amazon CloudFront?
   • Answer: CloudFront caches content at the edge (great for images/video), while Global Accelerator optimizes the network path using Anycast IPs (great for non-HTTP traffic like gaming or VoIP).
4. True or False: Availability Zones are connected to each other via low-latency, private fiber-optic networking.
   • Answer: True.

Comparison Tables