Using purpose-built AWS services for workloads

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

📘AWS Certified Solutions Architect – (SAA-C03)


1. What Are Purpose-Built AWS Services?

Purpose-built services are AWS services designed to solve a specific type of problem very efficiently.

Instead of building everything manually (like managing servers, databases, queues), AWS provides ready-made services that are:

  • Highly available (HA)
  • Fault-tolerant (FT)
  • Scalable
  • Managed (less operational work)

Key Idea for Exam:

Use the right tool for the job instead of building everything on EC2.


2. Why Use Purpose-Built Services?

Using these services helps you:

1. Improve Availability

  • Built-in replication
  • Multi-AZ support
  • Automatic failover

2. Improve Fault Tolerance

  • Service continues even if a component fails
  • No single point of failure

3. Reduce Operational Effort

  • No server management
  • No patching or maintenance

4. Scale Automatically

  • Handle sudden traffic increases

3. Categories of Purpose-Built AWS Services

For the exam, you must understand which service to use for which workload.


3.1 Compute Services

1. Amazon EC2

  • Virtual servers
  • Full control
  • Needs manual HA setup

👉 Use when:

  • Custom applications
  • Legacy systems

2. AWS Lambda

  • Serverless compute
  • Runs code automatically

👉 Use when:

  • Event-driven workloads
  • No server management required

HA/FT Benefits:

  • Auto scaling
  • Multi-AZ by default

3. Amazon ECS / EKS

  • Container management

👉 Use when:

  • Microservices
  • Containerized applications

3.2 Storage Services

1. Amazon S3

  • Object storage

👉 Use when:

  • Static files
  • Backups
  • Logs

HA/FT Features:

  • 99.999999999% durability
  • Multi-AZ storage

2. Amazon EFS

  • Shared file storage

👉 Use when:

  • Multiple EC2 instances need same files

3. Amazon FSx

  • Managed file systems (Windows, Lustre)

3.3 Database Services (Very Important for Exam)

1. Amazon RDS

  • Managed relational database

👉 Use when:

  • Structured data
  • SQL queries

HA Features:

  • Multi-AZ deployments
  • Automated backups
  • Read replicas

2. Amazon DynamoDB

  • NoSQL database

👉 Use when:

  • High performance at any scale

HA Features:

  • Serverless
  • Multi-AZ automatically

3. Amazon Aurora

  • High-performance relational DB

👉 Use when:

  • Need better performance than RDS

HA Features:

  • Auto-replication across AZs
  • Fast failover

3.4 Networking & Content Delivery

1. Elastic Load Balancer (ELB)

  • Distributes traffic

👉 Use when:

  • Multiple application servers

HA Benefit:

  • Removes single point of failure

2. Amazon CloudFront

  • Content Delivery Network (CDN)

👉 Use when:

  • Global content delivery

3. Amazon Route 53

  • DNS service

👉 Use when:

  • Traffic routing

HA Features:

  • Health checks
  • Failover routing

3.5 Messaging & Integration Services

1. Amazon SQS

  • Message queue

👉 Use when:

  • Decoupling applications

HA Features:

  • Highly durable
  • Prevents system failure due to overload

2. Amazon SNS

  • Notification service

👉 Use when:

  • Sending alerts/messages

3. Amazon EventBridge

  • Event routing

👉 Use when:

  • Event-driven architectures

3.6 Caching Services

Amazon ElastiCache

  • In-memory cache

👉 Use when:

  • Reduce database load
  • Improve performance

4. Key Exam Concept: Choose Managed Over Self-Managed

Example Scenario:

Instead of:

  • Running database on EC2 ❌

Use:

  • RDS / DynamoDB ✅

Why?

  • Built-in HA
  • Automatic backups
  • Failover handled by AWS

5. Common Design Patterns Using Purpose-Built Services


5.1 Decoupled Architecture

Use:

  • SQS between components

Benefit:

  • If one component fails → system continues

5.2 Event-Driven Architecture

Use:

  • Lambda + EventBridge

Benefit:

  • Automatic processing
  • No server management

5.3 Highly Available Web App

Use:

  • EC2 + ELB + Auto Scaling + RDS Multi-AZ

5.4 Serverless Architecture

Use:

  • API Gateway + Lambda + DynamoDB

6. Important Exam Tips

1. Prefer Serverless First

  • Lambda
  • DynamoDB
  • S3

2. Prefer Managed Services Over EC2

  • Less operational effort
  • Better reliability

3. Use Multi-AZ Services

  • RDS Multi-AZ
  • S3 (automatic)
  • DynamoDB (automatic)

4. Decouple Components

  • Use SQS, SNS

5. Avoid Single Point of Failure

  • Use load balancers
  • Use replication

7. Comparison Table (Very Important for Exam)

RequirementBest Service
File storageS3
Shared file systemEFS
Relational DBRDS / Aurora
NoSQL DBDynamoDB
Event processingLambda
Queue systemSQS
NotificationsSNS
Global contentCloudFront
DNS routingRoute 53
CachingElastiCache

8. Common Exam Questions (What AWS Expects You to Choose)

If question says:

  • “Fully managed” → Avoid EC2
  • “Automatic scaling” → Lambda / DynamoDB
  • “High availability with minimal effort” → Managed service
  • “Decouple application” → SQS
  • “Event-driven” → Lambda + EventBridge

9. Summary (Must Remember)

  • Purpose-built services are specialized AWS tools
  • They provide:
    • High availability
    • Fault tolerance
    • Scalability
  • Always:
    • Choose managed over self-managed
    • Avoid single points of failure
    • Use decoupling (SQS, SNS)
    • Use serverless where possible
Buy Me a Coffee