Task Statement 2.2: Design highly available and/or fault-tolerant architectures.

📘AWS Certified Solutions Architect – (SAA-C03)

1. What is AWS Global Infrastructure?

AWS Global Infrastructure is the worldwide network of data centers that AWS uses to run its cloud services.

It is designed to provide:

High availability (systems stay up)
Fault tolerance (systems keep working even if something fails)
Low latency (fast response time)
Scalability (handle growth easily)

2. Core Components of AWS Global Infrastructure

You must clearly understand these 3 main components:

2.1 AWS Regions

What is a Region?

An AWS Region is a geographical area that contains multiple data centers.

Examples:

us-east-1 (North Virginia)
eu-west-1 (Ireland)
ap-south-1 (Mumbai)

Key Points for Exam:

Each region is completely isolated from others
You choose a region when deploying resources
Data does NOT automatically move between regions
Each region has multiple Availability Zones

Why Regions Matter:

Used for disaster recovery
Helps meet compliance requirements
Reduces latency for users near that region

2.2 Availability Zones (AZs)

What is an Availability Zone?

An Availability Zone (AZ) is a separate data center (or group of data centers) within a region.

Each AZ:

Has independent power, cooling, and networking
Is physically separated from other AZs
Is connected with high-speed private network

Example:

Region → ap-south-1
AZs → ap-south-1a, ap-south-1b, ap-south-1c

Key Points for Exam:

AZs are designed for fault isolation
If one AZ fails, others still work
Used to build highly available systems

Important Design Principle:

👉 Always deploy applications across multiple AZs

2.3 Edge Locations

What are Edge Locations?

Edge locations are data centers used for content delivery and caching.

Used by services like:

Amazon CloudFront
AWS Global Accelerator

Key Points:

Located close to end users
Used to cache content
Improves performance and latency

3. Amazon Route 53 (DNS Service)

What is Route 53?

Amazon Route 53 is a DNS (Domain Name System) service.

It translates:

example.com → IP address

Key Features for Exam:

3.1 Highly Available DNS

Route 53 is globally distributed
Automatically routes traffic even if failures occur

3.2 Routing Policies (VERY IMPORTANT)

You must know these:

1. Simple Routing

One resource
No failover

2. Weighted Routing

Split traffic between multiple resources
Example: 70% → Server A, 30% → Server B

3. Latency-Based Routing

Sends users to lowest latency region

4. Failover Routing

Primary + Secondary setup
If primary fails → traffic goes to secondary

👉 Used for disaster recovery

5. Geolocation Routing

Routes based on user location

6. Multi-Value Routing

Returns multiple IPs
Improves availability

3.3 Health Checks

Route 53 can:

Monitor endpoints
Detect failures
Automatically redirect traffic

👉 Critical for fault tolerance

4. How AWS Global Infrastructure Enables High Availability

High availability means:
👉 System stays operational even if something fails

4.1 Multi-AZ Architecture

Design:

Deploy application in 2 or more AZs
Use load balancing

Example (IT Environment):

Web servers in AZ-A and AZ-B
Load balancer distributes traffic
If AZ-A fails → AZ-B handles traffic

4.2 Elastic Load Balancer (ELB)

Distributes traffic across multiple AZs
Automatically removes unhealthy instances

👉 Works closely with AZ design

4.3 Auto Scaling

Automatically adds/removes servers
Ensures capacity during failure or load increase

5. How AWS Global Infrastructure Enables Fault Tolerance

Fault tolerance means:
👉 System continues working even when failures occur

5.1 Multi-Region Architecture

Deploy application in multiple regions
Use Route 53 for routing

Example:

Primary region: ap-south-1
Secondary region: us-east-1

If primary fails → Route 53 redirects traffic

5.2 Data Replication

Services that support replication:

Amazon S3 (cross-region replication)
Amazon RDS (read replicas, multi-AZ)
DynamoDB (global tables)

6. Disaster Recovery Strategies (Exam Important)

Know these 4 types:

6.1 Backup and Restore

Backup data (e.g., S3)
Restore when failure happens

👉 Cheapest but slowest

6.2 Pilot Light

Minimal environment always running
Scale up during failure

6.3 Warm Standby

Smaller version of full system running
Quickly scale when needed

6.4 Multi-Site (Active-Active)

Full system in multiple regions
Traffic distributed across all

👉 Most expensive but highest availability

7. Exam Tips (VERY IMPORTANT)

1. AZ vs Region

AZ = high availability
Region = disaster recovery

2. Always Use:

Multiple AZs for production workloads
Load balancer + Auto Scaling

3. Route 53 Questions:

Failover → disaster recovery
Latency → performance optimization
Weighted → traffic splitting

4. Data Safety:

Use replication for fault tolerance
Use backups for recovery

5. Isolation Concept:

Regions are isolated
AZs are isolated but connected

8. Quick Summary

Component	Purpose
Region	Geographic isolation
Availability Zone	Fault isolation within region
Edge Location	Fast content delivery
Route 53	DNS + traffic routing
Multi-AZ	High availability
Multi-Region	Disaster recovery

Final Understanding

To pass the exam, remember:

High availability → Multiple AZs
Fault tolerance → Multi-region + replication
Traffic routing → Route 53
Performance → Edge locations
Design goal → No single point of failure