You can run ghuser.io's implementation anywhere. We are currently running it on AWS by making use of the following services.

• Services
  • AWS Lambda
  • AWS EC2
  • AWS EFS
  • AWS S3
  • AWS SQS
  • AWS Route 53
• Setting up AWS CLI

The most important part of ghuser.io is a Reframe web app. We deploy the web server on AWS Lambda using Apex/Up. This setup is best described by How to deploy a Reframe web app on Up.

A few AWS IAM adjustments are necessary:

• set up AWS CLI, and
• go to the apex-up/ subfolder.

The data of all profiles needs to be refreshed every day. For this we run a bot on an EC2 instance.

To create this EC2 instance:

• set up AWS CLI, and
• go to the ec2/ subfolder.

The data of all profiles is stored as a bare git repository on an EFS mounted on the EC2 instance.

To create this EFS:

• set up AWS CLI, and
• go to the efs/ subfolder.

The web app's front end running in the user's browser fetches the data of all profiles from S3. Our S3 bucket is a mirror of the data present on the EFS.

To create this S3 bucket:

• set up AWS CLI, and
• go to the s3/ subfolder.

The bot also processes profile requests, i.e. when a new user wants to get their profile. These requests are implemented as SQS messages.

To create this SQS queue:

• set up AWS CLI, and
• go to the sqs/ subfolder.

We bought the ghuser.io domain from Route 53. To see how this is set up:

• set up AWS CLI, and
• go to the dns/ subfolder.

In order to run the scripts present in these subfolders you first need to set up AWS CLI:

$ sudo pip install awscli==1.16.4
$ aws configure
AWS Access Key ID [None]: ********
AWS Secret Access Key [None]: ********
Default region name [None]: us-east-1
Default output format [None]: