AWS Solutions Architect Associate: A Study Guide

The AWS Certified Solutions Architect - Associate is AWS's general architecture certification at associate level, and it is often misunderstood. It is not a test of whether you can operate AWS. It is a test of design judgement: given a requirement, which combination of services meets it securely, reliably, quickly and cheaply. AWS's exam guide states that the exam validates the ability to design solutions based on the AWS Well-Architected Framework (AWS exam guide: SAA-C03).

Study for it as a design exam and it becomes manageable. Study it as a list of services to memorise and it does not.

The exam in numbers

Length, format and cost are from the certification page (AWS: Solutions Architect - Associate). As with other AWS exams, unanswered questions count as incorrect, and scoring is compensatory across sections. Check the page for current details before booking.

The exam guide's target candidate has at least one year of hands-on experience designing cloud solutions on AWS. The certification page adds that deep coding experience is not required, and that people with one to three years of IT experience have used it as a starting point.

The four domains

The domain names are four qualities of a good design, and they map closely to pillars of the Well-Architected Framework (AWS: Well-Architected pillars).

Security is the largest domain, and the last place to skimp.

How the questions work

Almost every question is a short scenario followed by a qualifier. The scenario describes a system and a need. The qualifier tells you which quality to optimise:

• "most cost-effective"
• "least operational overhead"
• "highest availability"
• "most secure"

Usually two or more of the options would work. Only one is best for the stated qualifier. A design that is correct but expensive is wrong when the question asks for cost-effective, and a design that is cheap but needs servers to manage is wrong when it asks for least operational overhead.

A reading method that helps:

1. Read the last sentence first to find the qualifier.
2. Underline the constraints in the scenario: data size, latency, durability, compliance, existing systems.
3. Eliminate options that break a constraint.
4. Choose between the survivors on the qualifier alone.

What to study in each domain

Secure (30%). IAM users, roles and policies; the principle of least privilege; cross-account access; VPC security groups and network ACLs; private connectivity to services; encryption at rest and in transit with KMS; secrets management. Read AWS IAM explained, VPC basics for data engineers and securing data on AWS with encryption and KMS.

Resilient (26%). Regions and Availability Zones; load balancing and auto scaling; decoupling with queues and events; multi-AZ databases and read replicas; backup and disaster recovery strategies and what each costs.

High-performing (24%). Choosing between S3, EBS and EFS; instance families; relational, key-value and warehouse databases; caching and content delivery; and, notably for data people, ingestion and transformation: streaming versus batch, Kinesis, Glue, and data lake patterns. S3 vs EBS vs EFS is typical of the comparisons asked.

Cost-optimised (20%). S3 storage classes and lifecycle rules; on-demand, reserved, savings plans and Spot purchasing; right-sizing; data transfer costs; serverless versus provisioned. S3 lifecycle policies covers one common topic.

Study by trade-off, not by service

The most effective preparation is a notebook of comparisons. For each pair, write when you would choose each and why:

When you can explain each row without notes, most scenario questions resolve themselves.

A study method

1. Read the exam guide, including the task statements in each domain.
2. Learn the core services hands-on: VPC, EC2, S3, IAM, RDS, a load balancer and auto scaling. Build a small two-tier application once, then delete it.
3. Build the comparison notebook as you go.
4. Read the Well-Architected Framework overview, enough to recognise each pillar's concerns.
5. Take practice questions, using AWS's official set first. For every wrong answer, find the constraint or qualifier you missed.
6. Sit the exam when practice results are steady.

How long this takes depends on experience. If you are new to IT, AWS suggests the Cloud Practitioner first; see the Cloud Practitioner guide.

Is it the right exam for data engineers?

It is a general architecture exam. Networking, compute and resilience take up much of it, and analytics services appear only as part of one domain. For someone whose work is pipelines and warehouses, the Data Engineer - Associate exam is the closer match.

It is still valuable. Data platforms sit inside VPCs, depend on IAM, and fail in the ways this exam teaches you to design against. Many data engineers take it first for that breadth, then the data exam. The AWS certification path for data roles discusses the order.

AWS data services and cloud architecture for data pipelines are covered in the Data Engineering course.

Sources

• AWS Certified Solutions Architect - Associate (SAA-C03) exam guide (AWS Certification Exam Guides) — AWS Documentation
• AWS Certified Solutions Architect - Associate (Amazon Web Services) — Amazon Web Services
• Content Domain 1: Design Secure Architectures (AWS Certification Exam Guides) — AWS Documentation
• Content Domain 3: Design High-Performing Architectures (AWS Certification Exam Guides) — AWS Documentation
• The pillars of the framework (AWS Well-Architected Framework) — AWS Documentation