Graduation Prediction System with AWS Autoscaling and Load Testing

This project demonstrates the deployment of a web application with autoscaling on AWS and load testing using Locust. The application predicts the graduation status of students based on their grades.

Table of Contents

• Project Overview
• Prerequisites
• Setting Up AWS Infrastructure
  • Launch EC2 Instance
  • Install Required Software on EC2
  • Deploy Streamlit Application
• Setting Up Load Balancer
  • Create Target Group
  • Create Load Balancer
  • Configure Security Groups
• Deploying Locust for Load Testing
  • Install Locust
  • Run Locust
  • Access Locust Web Interface
• Locust Configuration
• Monitoring and Autoscaling
• Contributing
• License

Project Overview

This project aims to demonstrate the implementation of autoscaling to improve the efficiency and performance of a web application during user traffic spikes. The web application is built using Streamlit and predicts student graduation status based on their grades. Locust is used for load testing.

Prerequisites

• AWS Account
• SSH Key Pair for EC2
• Basic knowledge of AWS services
• Python and pip installed on your local machine

Setting Up AWS Infrastructure

Launch EC2 Instance

1. Launch an EC2 instance:

   • AMI: Amazon Linux 2
   • Instance type: t2.micro (Free Tier eligible)
   • Key pair: Select an existing key pair or create a new one
   • Security group: Create a new security group with the following rules:
     • SSH (port 22)
     • HTTP (port 80)
     • Custom TCP (port 8501 for Streamlit)
     • Custom TCP (port 8089 for Locust)

2. Connect to your EC2 instance:

   ssh -i "path/to/your-key-pair.pem" ec2-user@your-ec2-public-dns
   

Install Required Software on EC2

Update packages and install dependencies:

sudo yum update -y
sudo yum install python3 git -y

Install pip and virtualenv:

sudo python3 -m ensurepip
sudo pip3 install --upgrade pip
sudo pip3 install virtualenv

Clone the project repository:

git clone https://github.com/Yoga-Syahputra/graduation-prediction-system.git
cd graduation-prediction-system

Set up a virtual environment:

virtualenv venv
source venv/bin/activate

Install project dependencies:

pip install -r requirements.txt

Deploy Streamlit Application

Run the Streamlit app:

nohup streamlit run app.py --server.port 8501 --server.headless true &

Verify the application:

Access the application at http://your-ec2-public-dns:8501

Setting Up Load Balancer

Create Target Group

1. Go to the EC2 Dashboard.
2. Under Load Balancing, select Target Groups.
3. Click on Create target group.
4. Select Instances as the target type.
5. Configure the target group:
   • Name: uat-target-group
   • Protocol: HTTP
   • Port: 8501
   • VPC: Select your VPC
6. Click Create.

Create Load Balancer

1. Go to the EC2 Dashboard.
2. Under Load Balancing, select Load Balancers.
3. Click on Create Load Balancer and select Application Load Balancer.
4. Configure the load balancer:
   • Name: uat-load-balancer
   • Scheme: Internet-facing
   • Listeners: HTTP (port 80)
   • Availability Zones: Select the same VPC and at least two subnets
5. Click Next: Configure Security Settings.
6. Click Next: Configure Security Groups, and select your security group.
7. Click Next: Configure Routing.
8. Configure the target group:
   • Target group: uat-target-group
   • Protocol: HTTP
   • Port: 8501
9. Click Next: Register Targets, and add your EC2 instance.
10. Click Next: Review, and then Create.

Configure Security Groups

1. Go to the EC2 Dashboard.
2. Under Network & Security, select Security Groups.
3. Select the security group associated with your EC2 instance.
4. Add the following inbound rules:
   • Type: HTTP, Protocol: TCP, Port range: 80, Source: 0.0.0.0/0
   • Type: Custom TCP, Protocol: TCP, Port range: 8501, Source: 0.0.0.0/0
   • Type: Custom TCP, Protocol: TCP, Port range: 8089, Source: 0.0.0.0/0

Deploying Locust for Load Testing

Install Locust

1. Install Locust:

   pip install locust

Run Locust

nohup locust -f locustfile.py --host http://your-load-balancer-url --web-host 0.0.0.0 &

Access Locust Web Interface

Access Locust interface:

• URL: http://your-ec2-public-dns:8089
• Configure the number of users and spawn rate, then start the test.

Locust Configuration

Configure number of users and spawn rate:

• Number of users: [adjustable]
• Spawn rate: [insert the number here] user per second
• Host: http://<load-balancer-url>

Monitoring and Autoscaling

• Monitor your application performance using the AWS Management Console.
• Set up CloudWatch Alarms to trigger scaling actions based on metrics like CPU utilization.
• Configure Autoscaling Groups to automatically add or remove instances based on demand.

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or bug fixes.

License

This project is licensed under the MIT License. See the LICENSE file for details.