📉 AWS 107: Save Money by Rightsizing - How to Change an EC2 Instance Type

Source: Dev.to

Hey Cloud Architects! 👋

Welcome to Day 7 of the #100DaysOfCloud Challenge: Change EC2 Instance Type! As the Nautilus DevOps team continues their migration, they’ve hit a common scenario: over‑provisioning. After monitoring their resources via KodeKloud Engineer, they realized one server wasn’t doing much work.

1️⃣ Introduction – What is Rightsizing? 💡

In the cloud you pay for what you provision. If a large server is running a tiny task, you’re wasting money.

• Rightsizing – Matching instance types and sizes to your workload’s performance and capacity requirements at the lowest possible cost.
• Vertical scaling – Changing an instance type is a form of vertical scaling; you’re essentially swapping the “hardware” specs (CPU/RAM) of your virtual server.
• Important requirement – You cannot change an instance type while the server is running; it must be stopped first.

2️⃣ Step‑by‑Step Guide – Scaling Down nautilus-ec2

Prerequisite: Ensure the instance has finished its Initializing phase and shows 2/2 status checks passed.

Step 2.1 – Stop the Instance

You can’t change the engine of a car while it’s driving – the instance must be stopped.

1. Log in to the AWS Console and navigate to the EC2 Dashboard.

   

2. Click on “Instances”.

   

3. Select the nautilus-ec2 instance.

   

4. Click “Instance state” → “Stop instance”.

   

5. Confirm by clicking “Stop”. Wait until the instance state changes to 🔴 Stopped.

   

(Continue with the remaining steps – change instance type, start the instance, verify the new configuration – in the original article.)

Step 2.2 – Change the Instance Type

Once the instance is stopped, the “Change instance type” option becomes available.

1. With the instance still selected, click the “Actions” button.

2. Navigate to “Instance settings” → “Change instance type.”

   

3. In the dropdown, search for and select t2.nano.

   

4. Click “Apply.”

Step 2.3 – Restart the Instance

Now that we’ve swapped the “hardware,” let’s turn the server back on.

1. With nautilus-ec2 selected, click “Instance state.”

2. Choose “Start instance.”

   

Step 2.4 – Verify the Change

• Wait for the instance state to return to 🟢 Running.

• Check the “Instance type” column in your dashboard – it should now display t2.nano.

• Ensure Status checks pass again.

  

3. Key Takeaways 📝

• Downtime is required: Changing an instance type requires stopping the instance, which means a brief period of downtime.
• Vertical scaling: The easiest way to handle a server that is either struggling (scale up) or under‑utilized (scale down).
• Compatibility: Ensure your AMI (Amazon Machine Image) and virtualization type are compatible with the new instance family you are moving to.

4. Common Mistakes to Avoid 🚫

• Forgetting to stop: The “Change instance type” option is greyed out if the instance is still running.
• Public IP change: Warning! If you are not using an Elastic IP, stopping and starting your instance will change its public IP address. Allocate an Elastic IP beforehand if you need a static address.
• Ignoring status checks: Never perform maintenance on an instance that is still “Initializing” or failing health checks.

5. Conclusion + Call to Action! 🌟

Rightsizing is a super‑power for DevOps engineers. By moving to a t2.nano, the Nautilus team is optimizing their cloud spend without sacrificing the performance needed for that specific task.

Are you keeping up with the 100 Days of Cloud Challenge?

💬 Let’s connect on LinkedIn: Let’s discuss cloud cost optimization!
👉 Hritik Raj

⭐ Support my journey on GitHub: Check out the logs for this task and more.
👉 GitHub – 100 Days of Cloud