Studying for the AWS Solution Architect Associate Exam

Studying for the AWS Solution Architect Associate Exam

I recently passed the AWS Solution Architect — Associate exam! In this article, I will share my two cents on the exam and any tips that I think you should consider when studying for this exam.

I took two steps to study for the exam.

Firstly, I took a course offered by A Cloud Guru to learn about each and every AWS service covered by the exam. I watched every lecture, did every lab and took notes from the classes.

Secondly, I did a ton of practice exams. I did the ones offered by A Cloud Guru, and any other free ones that I could find. I frequently checked my answers and researched on the ones that I got wrong.

First off, I took the A Cloud Guru (ACG) AWS Certified Solution Architect Associate course. This was the bulk of my study. I think it was an excellent course, it had just the right amount of theory and hands-on labs. The theory was taught in short videos which made the experience very engaging. The lectures also talk in this friendly and casual tone, which makes the classes fun as well. Any complex terminologies were explained in simple words. Overall, the course was very easy to follow.

In the past, I’ve tried courses where the theory was taught purely in text — it was so boring that I couldn’t finish the course. ACG’s course is the opposite. Its fun and engaging, which is exactly what you need when studying for a big exam like this one.

The hands-on labs in the courses were done in a small AWS environment where you can actually play around with the AWS console and services. The console was identical to the one you see in AWS. I had no technical issues when running the labs.

The labs are based on real-life scenarios, like “hosting a static website on a S3 bucket” or “setting up a VPC with an internet gateway”. This gave me an idea as to how AWS services could and should be used. The instructions for the labs were clear, but not too clear as to give away the answers. They also had a “challenge mode” available for those who wanted to do the labs without instructions.

Study Topics

The best thing about ACG’s course is how the study material is divided into service categories. I would recommend doing this even if you are not taking the ACG course. It helps when you are trying to get your head around the plethora of AWS services that is available today.

Here is how I would divide up the exam study:

  • Automation: Anything that automates your workflows on the Cloud. Things like EventBridge and Cloudformation
  • Big Data: Data analytics and machine learning. Sagemaker, Lakeformation, Redshift, EMR, etc
  • Compute: Things that run code. EC2 and Lambda
  • Containers: All things docker. ECS, ECR, Kubernetes
  • Cost Management: Services to help you manage AWS costs. Billing, Cost explorer are the main ones
  • Databases: There are many database services in AWS. The main ones are RDS, Aurora, DynamoDB
  • Decoupling: Services you use between a server and another server. Think SQS and SNS
  • Management & Governance: Where you do admin stuff. IAM, Organizations, Control Tower
  • Migration: Services you use when migrating from on-prem to cloud. XXX Migration Service, Datasync, the Snow series, etc
  • Monitoring: Where logs and metrics are stored. Things like CloudWatch, CloudTrail
  • Networking: VPC and everything beneath it
  • Security: There are many services here. Security Groups, firewalls, Shield, Inspector, GuardDuty
  • Storage: Where you store data. S3, EFS, EBS

Once I finished the A Cloud Guru course, I moved on to practice exams. I did a ton of these.

The first point I’d like to make is this: don’t be afraid of looking at the answers! When studying for an exam, a lot of people try and avoid looking at the answers until they finish the whole exam. Some online practice exams are formatted like this as well.

When I study, I prefer to look at each question, make an educated guess and compare the answers before I move on to the next question. If I get a question wrong, I make sure to go over the materials and understand why I went wrong. This way, the thought process for the question is still fresh in my mind. If I’m reviewing answers at the very end of a 65-question exam, not only am I mentally exhausted, but I won’t remember why I chose the wrong answers. I like to have small and frequent feedback loops for my studies.

Also, there are quite a few trick questions in the exam. To avoid getting tricked, make sure to extract key words when answering a question. In particular, big hints are at the end of the question, where it describes how they want to perform a job. For example, if the question ends with “most cost effectively”, you will prioritize choices that are cheap, such as EC2 spot instances and serverless options. If the question is about migrating an on-prem monolithic application and it asks for the “least amount of code changes”, you wouldn’t try and redesign the whole application, but rather look at running the same code on an EC2 instance.

My pick of services to look out for

Here are 5 services that AWS loves and tends to be an answer in the exam:

  1. DynamoDB — Its got all the buzzwords in it. NoSQL, infinitely scalable, it even has a serverless offering now too!
  2. VPC — This service is way too big. You need to learn some networking if you want to pass this exam.
  3. Lambda — Does this need an introduction? If you want to run code easily, use Lambda
  4. CloudFront — It’s so much more than a content delivery service. It can do failover, serving content, edge computing, cook your dinner, fix your car.
  5. SQS — For some reason AWS loves this thing. If SQS is one of the choices, I can guarantee SQS is the answer for your question. At least in 2024.

Conclusion

That’s a wrap! Study hard, prep hard and good luck with your AWS certification journey :)

Read more

【ほらな】Claude Codeが凄いよって話【言っただろ】

【ほらな】Claude Codeが凄いよって話【言っただろ】

はじめに 去年、私はLLM時代のプログラミングについてというブログ記事を公開しました。 LLM時代のプログラミングについてChatGPTなどのLLM(大規模言語モデル)が世界を席巻してからもうすぐ二年が経とうとしています。文章執筆はもちろん、「翻訳」「情報収集」「悩み相談」「今夜の献立」「ぼくのかんがえたさいきょうのかのじょ」などなど…LLMは、その驚異的な応用性をもって、多岐にわたる分野で使われるようになりました。 僕が生計を立てている「プログラミング」界隈も、ここ二年でガラッと変化しました。LLMは、名前の大規模「言語」モデルにある通り、テキストに特化した機械学習モデルです。コードの生成なんてチョチョイのチョイです。 また、LLMの精度も急速に向上しています。OpenAIやAnthropicなどのAI会社は、まるで雨後のタケノコのようにポンポンと新作LLMを発表。現在のフラグシップであるGPT-4oやClaude3.5-Sonnetの精度は半端じゃなく、綺麗なWebアプリもモデルに要件を伝えるだけで作れちゃいます。 ぶっちゃけ、LLMの精度がこんなにも向上すると、自分でプログラ

By Roland Thompson
zshの模様替えをしました

zshの模様替えをしました

💡この記事は下の動画の受け売りです: はじめに 今年も早いもので、すでに三月。春の始まりです。 春といえば新しい生活。心機一転! 周りの環境が大きく変わる方も多いのではないでしょうか? 今年の春、私の人生に大した変化はないのですが、それでも人生にちょっと新しい風を吹かしたいな~と思うので、今回思い切って模様替えをしようと思います! ターミナル環境のね。 zshellって? シェル (shell) とは、OSとユーザーの仲介役。ターミナルを開いたときに動くプログラムがシェルです。 そのシェルにも色々あるんです。sh, bash, fish, ksh などなど…。今回は、その中のzsh というシェルの話。 zshは高いカスタマイズ性が人気です。zsh そのものの機能と豊富なプラグインでターミナル環境を自分の好きなように設定できます。 私もzshを使い始めて数年たちますが、欲しいところにちゃんと手が届く、という印象です! …とか言いつつ私は面倒くさがりなので、自分で.zshrc をイジることはしません。YouTubeやSNSなどで見つけたカッコいいタ

By Roland Thompson
Terraformを使って"AWS Lambdaとその取り巻き"を召喚しよう

Terraformを使って"AWS Lambdaとその取り巻き"を召喚しよう

はじめに 人生、いろんな「派閥」ってありますよね。私が属するITインフラ業界にも色々あります。どのクラウドを使うか、どのIDEを使うか、どのOSを使うか…枚挙に暇がありません。 インフラ(IaC)言語もその一つ。私はこの業界に足を踏み入れてから、ずっとCloudFormation派閥です。私はAWS専門だったので、AWS公式のCloudFormationで仕事が成り立ってました。 しかし最近、AzureやDatabricks関連の仕事も私に降ってくるようになりました。そうなると、AWS限定のCloudFormationでは対応できません。 そんなとき、複数のクラウドプラットフォームに対応できるTerraformという存在を耳にしました。 Terraformの練習として、AWS Lambdaとその取り巻き(ECR, IAM, Secrets Manager, CloudWatch, SQS) を召喚してみたので、この記事にまとめます。 環境構築 まずは公式マニュアルを参考にTerraformをインストールしましょう。 ターミナルからTerraformが動けば、インストー

By Roland Thompson