Task Statement 2.1: Design scalable and loosely coupled 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 efficiently.
Instead of building everything from scratch (for example, managing servers or databases manually), AWS provides services that are already optimized for specific workloads.
Key Idea:
Choose the right tool for the job instead of using one service for everything.
2. Why Use Purpose-Built Services?
Benefits (Important for Exam)
1. Better Performance
- Optimized for specific workloads
- Example: Database queries run faster on a database service than on general compute
2. Scalability
- Automatically scales based on demand
- No need for manual intervention
3. Lower Operational Overhead
- AWS manages infrastructure, patching, backups
4. Cost Efficiency
- Pay only for what you use
- Avoid over-provisioning
5. Built-in High Availability
- Many services are multi-AZ by default
3. Categories of Purpose-Built Services
For the exam, you must understand which service to choose based on workload type.
3.1 Compute Services
Used to run applications.
Amazon EC2
- General-purpose virtual servers
- Full control over OS
- Use when:
- You need customization
- You manage the infrastructure
AWS Lambda
- Serverless compute
- Runs code without managing servers
- Use when:
- Event-driven workloads
- Short-running tasks
AWS Fargate
- Serverless containers
- No need to manage servers
Exam Tip:
- If no server management required → Lambda or Fargate
- If full control needed → EC2
3.2 Storage Services
Used to store data.
Amazon S3
- Object storage
- Highly scalable and durable
- Use for:
- Static files
- Backups
- Data lakes
Amazon EBS
- Block storage for EC2
- Use when:
- Low-latency disk storage is required
Amazon EFS
- Shared file system
- Use when:
- Multiple instances need access to same files
Exam Tip:
- Static data → S3
- OS or database disk → EBS
- Shared storage → EFS
3.3 Database Services (Very Important)
Choosing the correct database is a frequent exam question.
Relational Databases
Amazon RDS
- Managed relational database
- Supports MySQL, PostgreSQL, etc.
- Use when:
- Structured data
- Complex queries (JOINs)
Amazon Aurora
- High-performance relational DB
- Better scalability than RDS
NoSQL Databases
Amazon DynamoDB
- Key-value and document database
- Fully serverless
- Use when:
- Massive scale
- Low latency
- No complex joins
Specialized Databases
Amazon ElastiCache
- In-memory caching
- Use when:
- Need faster reads
Amazon Redshift
- Data warehouse
- Use when:
- Analytics and reporting
Amazon Neptune
- Graph database
- Use when:
- Relationships between data are important
Amazon Timestream
- Time-series database
- Use when:
- Data changes over time (metrics, logs)
Exam Tip:
- Transactions → RDS/Aurora
- High-scale NoSQL → DynamoDB
- Caching → ElastiCache
- Analytics → Redshift
3.4 Messaging and Integration Services
Used for decoupling systems.
Amazon SQS
- Message queue
- Ensures reliable message delivery
Amazon SNS
- Pub/Sub service
- Push messages to multiple subscribers
Amazon EventBridge
- Event bus
- Routes events between services
Exam Tip:
- Queue → SQS
- Fan-out → SNS
- Event routing → EventBridge
3.5 Analytics Services
Used to process large data.
Amazon Athena
- Query data in S3 using SQL
AWS Glue
- ETL (Extract, Transform, Load)
Amazon Kinesis
- Real-time data streaming
Exam Tip:
- Query S3 → Athena
- Data transformation → Glue
- Streaming → Kinesis
3.6 Machine Learning Services
Amazon SageMaker
- Build, train, deploy ML models
3.7 Networking and Content Delivery
Amazon CloudFront
- Content Delivery Network (CDN)
Elastic Load Balancer (ELB)
- Distributes traffic
4. How to Choose the Right Purpose-Built Service
For the exam, always analyze:
Step 1: Identify Workload Type
- Compute?
- Storage?
- Database?
- Messaging?
Step 2: Identify Requirements
- Scalability
- Latency
- Data structure
- Cost
Step 3: Choose the Best Service
5. Common Exam Scenarios
Scenario 1: High-traffic web application
- Compute → EC2 / Lambda
- Storage → S3
- Database → DynamoDB or Aurora
Scenario 2: Real-time data processing
- Streaming → Kinesis
- Processing → Lambda
- Storage → S3
Scenario 3: Decoupled microservices
- Messaging → SQS
- Notifications → SNS
Scenario 4: Analytics workload
- Storage → S3
- Query → Athena
- ETL → Glue
6. Key Comparison Table
| Workload Type | Service |
|---|---|
| Object storage | S3 |
| Block storage | EBS |
| File storage | EFS |
| Relational DB | RDS / Aurora |
| NoSQL DB | DynamoDB |
| Cache | ElastiCache |
| Data warehouse | Redshift |
| Queue | SQS |
| Pub/Sub | SNS |
| Event routing | EventBridge |
7. Important Exam Tips
1. Avoid Using EC2 for Everything
- AWS prefers managed services
2. Choose Serverless When Possible
- Lambda, DynamoDB, S3 are often correct answers
3. Use Specialized Databases
- Do not use relational DB for everything
4. Think About Scaling
- Services like DynamoDB and S3 scale automatically
5. Decouple Components
- Use SQS, SNS, EventBridge
8. Summary
- Purpose-built services are optimized for specific workloads
- They improve:
- Performance
- Scalability
- Cost efficiency
- Always choose services based on:
- Workload type
- Requirements
- The exam tests your ability to:
- Select the right AWS service for the right use case
