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

📘AWS Certified Solutions Architect – (SAA-C03)

1. What is Load Balancing?

Load balancing is the process of distributing incoming network traffic across multiple servers (resources) so that:

No single server is overloaded
Applications remain highly available
Performance is consistent
Failures do not impact users

In AWS:

Load balancing is provided by Elastic Load Balancing (ELB).

2. Why Load Balancing is Important

Load balancing is essential for:

1. High Availability

Traffic is spread across multiple servers (often in different Availability Zones)
If one server or AZ fails, others continue serving traffic

2. Fault Tolerance

Unhealthy servers are automatically removed
Traffic is routed only to healthy targets

3. Scalability

Works with Auto Scaling Groups
Automatically adjusts as traffic increases or decreases

4. Better Performance

Prevents overload on individual instances
Improves response time

3. Types of AWS Load Balancers

AWS provides 4 types of load balancers:

1. Application Load Balancer (ALB)

Works at Layer 7 (HTTP/HTTPS)
Best for web applications
Supports advanced routing

2. Network Load Balancer (NLB)

Works at Layer 4 (TCP/UDP)
Ultra-high performance, low latency

3. Gateway Load Balancer (GWLB)

Used for security appliances
Works with firewalls and inspection systems

4. Classic Load Balancer (CLB)

Legacy (not recommended for new architectures)

👉 For the SAA-C03 exam, ALB is the most important.

4. Application Load Balancer (ALB) – Deep Dive

4.1 What is ALB?

Application Load Balancer (ALB) distributes HTTP/HTTPS traffic based on content of the request.

It can make routing decisions based on:

URL path
Hostname
Headers
Query strings

4.2 Key Features of ALB

1. Layer 7 Routing (Content-Based Routing)

ALB can route traffic based on request details.

Examples (IT-focused):

/api/* → Backend API servers
/images/* → Image processing servers
admin.example.com → Admin application

2. Target Groups

ALB does not send traffic directly to instances. Instead, it sends traffic to Target Groups.

A target group contains:

EC2 instances
ECS containers
IP addresses
Lambda functions

👉 Each target group is associated with:

Health checks
Routing rules

3. Listeners

A listener checks for incoming traffic.

Defined by protocol + port
- HTTP : 80
- HTTPS : 443

Each listener has rules to route traffic.

4. Listener Rules

Rules define how traffic is routed.

Each rule has:

Condition (e.g., path, host)
Action (forward to target group)

👉 Rules are evaluated in order (priority-based).

5. Health Checks

ALB continuously checks whether targets are healthy.

If a target fails → it is removed from routing
If it recovers → it is added back

Health checks use:

HTTP/HTTPS endpoints
Example: /health

6. Multi-AZ Support

ALB automatically distributes traffic across:

Multiple Availability Zones

👉 This ensures:

High availability
Fault tolerance

7. Integration with Auto Scaling

ALB works with Auto Scaling Groups (ASG):

New instances → automatically registered
Terminated instances → automatically removed

8. SSL/TLS Termination

ALB can handle HTTPS:

Stores SSL certificates (via AWS Certificate Manager)
Decrypts traffic before forwarding

👉 This reduces load on backend servers

9. WebSocket and HTTP/2 Support

Supports modern web applications
Useful for real-time communication systems

10. Sticky Sessions (Session Affinity)

Ensures a user is sent to the same backend server
Uses cookies

👉 Useful for applications that store session data locally

5. How ALB Works (Step-by-Step Flow)

User sends request to ALB DNS name
Listener receives request (HTTP/HTTPS)
Listener evaluates rules
Request is forwarded to a target group
ALB selects a healthy target
Target processes request and responds

6. ALB vs NLB (Exam Comparison)

Feature	ALB	NLB
Layer	Layer 7	Layer 4
Protocols	HTTP/HTTPS	TCP/UDP
Routing	Content-based	IP/port-based
Performance	Moderate	Very high
Use case	Web apps	High-performance apps

👉 Exam tip:

If question mentions HTTP routing or URL-based routing → ALB
If question mentions ultra-low latency or TCP → NLB

7. Important Concepts for the Exam

7.1 Cross-Zone Load Balancing

ALB distributes traffic evenly across all targets in all AZs
Enabled by default

7.2 Idle Timeout

Time ALB waits before closing inactive connection

7.3 Deregistration Delay

Time given to complete in-flight requests before removing a target

7.4 Security

ALB uses:

Security Groups (unlike NLB)
Can restrict:
- Source IP
- Ports

8. Common Architecture Patterns (Exam-Focused)

1. ALB + Auto Scaling

Dynamic scaling
High availability

2. ALB + ECS (Containers)

Microservices architecture
Each service → different target group

3. ALB + Lambda

Serverless backend
Event-driven processing

9. When to Use ALB (Exam Clues)

Choose ALB when you see:

HTTP/HTTPS workloads
Microservices architecture
Path-based routing
Host-based routing
Web applications
Need for WebSockets or HTTP/2

10. When NOT to Use ALB

Avoid ALB when:

Need ultra-low latency → use NLB
Need static IP → use NLB
Non-HTTP protocols → use NLB

11. Common Exam Traps

Trap 1:

“Route based on URL path”
→ Correct answer: ALB

Trap 2:

“Handle millions of requests per second with low latency”
→ Correct answer: NLB

Trap 3:

“Use Lambda as target”
→ Only supported by ALB

Trap 4:

“Need security group on load balancer”
→ Only ALB supports this (NLB does not)

12. Quick Summary (Revision)

Load balancing distributes traffic across servers
AWS uses Elastic Load Balancing (ELB)
ALB = Layer 7, HTTP/HTTPS, smart routing
Key components:
- Listener
- Rules
- Target Groups
Supports:
- Path-based routing
- Host-based routing
- Health checks
- Auto Scaling integration
Default choice for web applications