Launching and Connecting to a Free AWS EC2 Instance for NASA Earthdata Cloud Access

This guide shows you how to:

1. Launch a free Amazon EC2 virtual server in the cloud (Free Tier eligible)
2. Securely connect to it from your computer
3. Prepare the instance for working with NSIDC DAAC Earthdata Cloud data
4. Transfer files between your computer and the instance

No prior cloud experience needed — this walkthrough is designed for first-time users!

Prerequisites

• An AWS account
• A NASA Earthdata Login account
• A terminal application:
  • macOS/Linux → Terminal
  • Windows → Git Bash or WSL

Part 1: Launch a Free EC2 Instance

Cost Warning: Following the steps below with the specified Free Tier options will not incur charges. Using larger instance types, extra storage, or additional services may incur costs.

1.   Log in to the AWS Console

• Go to AWS Management Console and sign in.

2.   Set the Correct AWS Region

• In the Region selector (upper-right corner), choose: United States (Oregon) — us-west-2
• Ensures all resources are in the same region and avoids unexpected cross-region transfer costs.

3.   Open the EC2 Dashboard

 Option 1: Search Bar

• Type EC2 in the search bar at the top → select EC2 under Services.

 Option 2: All Services Menu

• Click All services in the left navigation menu → Under the Compute category, select EC2.

4.   Launch a New Instance

• Click Launch instance (or go to Instances in the left menu — the button is in the upper-right corner)
• Use the following settings:

Important: Place your .pem key file into a secure folder — you’ll need it to connect later.

Optional Recommendations

• Add a security group rule for port 8888 if you plan to run JupyterLab. Guide here: https://nsidc.org/data/user-resources/help-center/accessing-nasa-earthdata-cloud-jupyterlab-ec2
• Increase storage if working with large datasets (may incur costs).
• Explore other AMIs or instance types (check Free Tier eligibility to avoid charges).

5.   Confirm the Instance is Running

• Click Launch instance
• Go to the Instances panel (left navigation → Instances)
• Confirm the Instance state column shows Running before continuing

Stop or Terminate Your Instance

Cost Note:
Stopped instances: no compute charges, but storage may incur small costs.
Terminated instances: no charges, data is deleted.

To Stop or Terminate:

1.   In the AWS Management Console, go to EC2 → Instances.

2.   Select your instance.

3.   Click the Instance state dropdown.

4.   Choose Stop instance or Terminate instance.

5.   Confirm your choice.

Part 2: Connect to Your EC2 Instance

Important: You’ll be connecting to a remote server. Commands you run in this terminal will affect your EC2 instance, not your local computer.

1.   Select Your Instance

• On the Instances page (where you landed after launching), check the box next to your new instance (e.g., my_earthdata_ec2)

2.   Open the Connect Panel

• At the top right of the Instances table, click Connect (located to the left of Instance state and Actions).
• Select the SSH client tab. This shows the SSH command you will use to connect.

3.   Open a Terminal on Your Computer

• macOS/Linux: Open Terminal
• Windows: Open Git Bash or WSL

4.   Prepare Your .pem Key File

Your .pem file is the secure key that lets you log in. Its permissions must be restricted.

macOS/Linux

Navigate to the folder containing your .pem key life and run:

chmod 400 "<your_key>.pem"

This ensures only you can read the key, which SSH requires.

Windows

     1.   Locate your .pem file in File Explorer.

     2.   Right-click → Properties → Security → Advanced

     3.    Disable inheritance:

            Click Disable inheritance

            Select Convert inherited permissions into explicit permissions.

     4.   Remove unnecessary users/groups:

            Select SYSTEM, Administrators, Authenticated Users, and Everyone (if present), and click Remove

     5.   Keep only your primary user account and ensure it has Full control.

     6.   Click OK to apply changes.

            Your .pem file is now secured, equivalent to chmod 400.

5.   Copy the SSH Command

From the SSH client tab, copy the example command. It usually looks like this:

ssh -i "<your_key>.pem" ubuntu@<your_instance_public_dns>

• Replace <your_key> with your .pem key filename
• Replace <your_instance_public_dns> with your instance’s Public IPv4 DNS (found in the Connect panel)

6.   Connect to Your Instance

• Paste the command into your terminal and press Return.
• If prompted with Are you sure you want to continue connecting?, type yes and press Return
• If successful, your prompt will change to show the remote EC2 instance (e.g., ubuntu@ip-172-31-xx-xx:~$)

Note: From this point forward, your terminal is connected directly to your EC2 instance in the cloud. Any commands you type will run on that remote machine, not on your local computer. This is your EC2 terminal.

Part 3: Prepare Your EC2 Instance for Cloud Access

3.1 (Optional) Install AWS CLI

Cost Note: Installing AWS CLI is free. Data transfer from S3 may incur costs depending on data set size.

You only need this step if you want to download cloud-hosted data directly from S3. Otherwise, skip to 3.2.

1.   Download the AWS CLI v2 installer (Linux x86_64):

curl "<https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip>" -o "awscliv2.zip"

2.   Unzip the installer (install unzip if needed):

sudo apt update && sudo apt install -y unzip
unzip awscliv2.zip

3.   Run the installer:

sudo ./aws/install

4. Verify installation:

aws --version

You should see output like: aws-cli/2.x.x Python/3.x.x Linux/x86_64

Tip: To update AWS CLI later, just repeat the above steps with the latest installer.

3.2 Set Up NASA Earthdata Access

You’ll need to configure your EC2 instance so it can authenticate with NASA Earthdata services.

1.   Create a .netrc file with your Earthdata Login

echo "machine urs.earthdata.nasa.gov login your_username password your_password" >> ~/.netrc
chmod 600 ~/.netrc

• Replace your_username and your_password with your NASA Earthdata Login credentials.
• The .netrc file stores your login information securely so that Earthdata services can verify your requests.

Windows users: This step is different on Windows. Follow this guide for creating a .netrc file on Windows: https://nsidc.org/data/user-resources/help-center/creating-netrc-file-earthdata-login

2.   (Optional) Configure AWS temporary credentials

If you installed AWS CLI, export your temporary AWS S3 credentials:

export AWS_ACCESS_KEY_ID=<your_accessKeyId>
export AWS_SECRET_ACCESS_KEY=<your_secretAccessKey>
export AWS_SESSION_TOKEN=<your_sessionToken>

• Replace <your_accessKeyId>, <your_secretAccessKey>, and <your_sessionToken> with your actual AWS credentials.
• Get your temporary credentials here: NSIDC S3 credentials
• These credentials expire after a few hours, so you’ll need to refresh them for each new session.
• Confirm your AWS credentials (optional, if using AWS CLI): aws sts get-caller-identity. This will return your AWS account identity if the credentials are valid.

Part 4: Transfer Files To and From Your EC2 Instance

You can move files between your local computer and your EC2 instance using scp (secure copy) command.

Important: These commands must be run in a local terminal, not in the EC2 terminal. The EC2 terminal is connected to the virtual machine itself.

4.1 Upload a File to EC2

To send a file (e.g., a text file containing S3 URLs) from your computer to your EC2 instance:

scp -i <your-key>.pem local_file.txt ubuntu@<your-public-dns>:/home/ubuntu/

• Replace <your-key>.pem with your SSH key file.
• Replace local_file.txt with the file you want to upload.
• Replace <your-public-dns> with your instance’s Public DNS (found in the Connect panel in Part 2).

4.2 Download Files from EC2

To copy files from EC2 back to your local machine:

Download all files in /home/ubuntu/

scp -i <your-key.pem> -r ubuntu@<your-public-ip>:/home/ubuntu/ .

• -r = recursive (downloads folders and files inside).
• . = your current local directory.

Download only .h5 files

scp -i my-key.pem "ubuntu@<your-public-ip>:/home/ubuntu/*.h5" .

• Quotes prevent your local shell from expanding *.h5 before sending the command.
• This ensures the wildcard expansion happens on the EC2 instance rather than on your local machine.
• You can use this wildcard technique to target any specific file extensions you want to download, not just .h5 files.

Part 5: (Optional) Download Data from S3 on EC2

If you have AWS CLI installed and configured on your EC2 instance, you can download multiple S3 files efficiently.

1.   Upload a text file containing a list of S3 URLs to your EC2 instance (see Part 4.1)

2.   On the EC2 terminal, run the following script to download each file:

while read -r s3url; do
    echo "Downloading $s3url"
    aws s3 cp "$s3url" .      # Assumes credentials are configured and valid
done < S3-URLs.txt

• Replace S3-URLs.txt with the name of your text file.
• The script will download each file listed in the text file to the current directory on your EC2 instance.

Cost Note: S3 downloads from NASA buckets are usually free. Large downloads may consume bandwidth. Check dataset policies.

Tip: This is a quick way to verify your EC2 instance is configured correctly for cloud access and can handle batch downloads.

Part 6: Beginner-Friendly Ideas

Once your EC2 instance is ready, try the following to become comfortable with your cloud environment:

• Explore Linux commands: ls, cd, mkdir, rm
• Test file transfers with scp
• Edit files directly on EC2 with nano or vim
• Run Python scripts or Jupyter notebooks

Final Thoughts

You now have a fully functional, Free Tier EC2 environment ready for cloud-based Earth science workflows. This setup gives you:

• Secure access to your EC2 instance via SSH
• File transfer capabilities between your computer and EC2
• Access to NASA Earthdata Cloud and S3 data
• A reproducible workflow for Python scripts or Jupyter notebooks

With this foundation, you can safely explore NASA Earthdata collections and scale up worflows if needed — always checking Free Tier eligibility to avoid unexpected costs.