Task Statement 4.2: Design cost-optimized compute solutions.

📘AWS Certified Solutions Architect – (SAA-C03)

This topic is very important for the AWS Solutions Architect Associate exam, especially under cost optimization. The goal is simple:

👉 Use compute resources efficiently so you don’t waste money.

Instead of running large servers all the time, AWS provides smarter ways to run applications:

Containers
Serverless computing
Microservices architecture

Let’s understand each one in a simple and exam-focused way.

1. What is Compute Utilization Optimization?

Compute utilization means:

👉 How effectively your application uses CPU, memory, and compute resources

Poor Utilization (Bad for cost ❌)

EC2 instance running at 10% CPU
Server idle most of the time
Over-provisioned resources

Good Utilization (Cost-efficient ✅)

Resources used close to capacity
Auto-scaling based on demand
No idle compute

2. Containers (Efficient Resource Sharing)

What are Containers?

Containers package:

Application code
Dependencies
Runtime

👉 They run multiple applications on the same EC2 instance efficiently.

AWS Services for Containers

Amazon ECS (Elastic Container Service)
Amazon EKS (Elastic Kubernetes Service)
AWS Fargate (serverless containers)

Why Containers Improve Utilization

Traditional EC2 Model

One application per VM
Wastes CPU and memory

Container Model

Many containers share same OS
Better CPU & memory usage

Example (IT Scenario)

A company runs:

Web app
API service
Background worker

Instead of:

3 EC2 instances ❌

They run:

3 containers on 1 or 2 instances ✅

👉 Less cost, better utilization

Key Benefits (Exam Points)

High resource utilization
Faster deployment
Lightweight compared to VMs
Easy scaling
Portable

Cost Optimization Tip

👉 Use AWS Fargate when:

You don’t want to manage servers
Workloads are variable

👉 Use ECS/EKS on EC2 when:

You want more control and lower cost at scale

3. Serverless Computing (No Idle Resources)

What is Serverless?

You don’t manage servers at all.

👉 AWS runs code only when needed

Main AWS Serverless Services

AWS Lambda (core service)
Amazon API Gateway
AWS Step Functions
Amazon EventBridge

How It Optimizes Cost

Traditional EC2

Running 24/7 → paying even when idle ❌

Serverless (Lambda)

Runs only when triggered ✅
Charged per execution

Example (IT Scenario)

A system processes:

File uploads occasionally

Using EC2:

Server runs all day ❌

Using Lambda:

Runs only when file arrives ✅

👉 Zero idle cost

Key Benefits (Exam Points)

No server management
Automatic scaling
Pay per use
Built-in high availability

Limitations (Important for Exam)

Execution time limits
Cold starts (latency)
Not ideal for long-running processes

4. Microservices Architecture (Efficient Scaling)

What are Microservices?

Application is divided into small independent services.

Instead of one big app (monolith), you have:

Auth service
Payment service
Notification service

Why Microservices Improve Utilization

👉 Each service scales independently

Monolithic Application ❌

Entire app scales even if one part is busy
Wastes resources

Microservices Architecture ✅

Only busy service scales
Saves cost

Example (IT Scenario)

An application has:

Login service (low traffic)
Image processing service (high traffic)

With microservices:

Scale only image service
Keep login service small

👉 Better utilization

AWS Services Used

AWS Lambda (serverless microservices)
Amazon ECS / EKS (containerized microservices)
Amazon API Gateway
AWS Step Functions

Key Benefits (Exam Points)

Independent scaling
Fault isolation
Faster deployments
Better resource efficiency

5. Combining All Three (Very Important for Exam)

Modern architectures combine:

Containers + Microservices

Each microservice runs in a container

Serverless + Microservices

Each microservice is a Lambda function

Example Architecture

API Gateway → Lambda (microservices)
Or:
Load Balancer → ECS/EKS (containers)

👉 This leads to:

Maximum utilization
Minimum cost

6. Auto Scaling (Must Know)

Even with containers or EC2:

👉 Use Auto Scaling to match demand

Benefits

Scale out when traffic increases
Scale in when traffic decreases
Avoid over-provisioning

7. Exam Tips (Very Important)

When to Choose Containers

Multiple applications sharing resources
Need portability
Long-running workloads

When to Choose Serverless

Event-driven workloads
Unpredictable traffic
Short execution tasks

When to Choose Microservices

Large applications
Independent scaling needed
Frequent deployments

8. Common Exam Scenarios

Scenario 1

Application has unpredictable traffic
👉 Use: AWS Lambda (serverless)

Scenario 2

Multiple services need to run efficiently on same infrastructure
👉 Use: Containers (ECS/EKS)

Scenario 3

Only one part of application needs scaling
👉 Use: Microservices architecture

Scenario 4

Avoid paying for idle servers
👉 Use: Serverless

9. Key Takeaways

Containers → maximize resource usage on servers
Serverless → eliminate idle compute cost
Microservices → scale only what is needed
Auto Scaling → match supply with demand

👉 Final goal:
Use only the compute you actually need — nothing more, nothing less