AN APPROACH FOR PREVENTING OVERLOAD OF THE EC2 SERVICE

Abstract

The aim of this article is to present an approach for preventing service overload in Elastic Compute Cloud (EC2) within Amazon Web Services (AWS). The study analyzes methods for preventing overload of virtual machines in AWS EC2 by comparing vertical and horizontal scaling. Developed block diagrams are presented to visually illustrate the architecture and processes involved in dynamic resource management. A practical implementation of horizontal scaling of virtual machines is also provided. For this purpose, a Python script using the boto3 library has been developed, which fully automates the configuration of key cloud components such as Launch Template, Auto Scaling Group, and Scaling Policy. The proposed solution minimizes the need for manual intervention, protects cloud-based systems from failures during peak loads, and ensures the development of a resilient, flexible, and cost-effective infrastructure. Compared to the manual configuration through the AWS console, the proposed approach groups all required steps into a single Boto3 script that creates the Launch Template, the Auto Scaling Group and the Scaling Policy in one run. The block diagrams visualise both scaling workflows in a unified form, and the scripted procedure can be executed on a schedule through cron, which makes the solution suitable for use in the training of cadets and students at the Nikola Vaptsarov Naval Academy.