A Hybrid Monitoring Framework for Detecting and Blocking Cross-VM Network Attacks in Cloud Infrastructure

Abstract

This project presents a cloud-based file management system integrated with a virtual machine (VM) monitoring mechanism to identify and handle abnormal user activities. The system allows users to upload and download files through a simulation interface while a monitoring component continuously analyses incoming requests. Normal requests are processed and stored successfully, whereas suspicious or unusually large requests are detected as potential attacks and are blocked to protect system resources. The implementation demonstrates how real-time monitoring can prevent system crashes caused by malicious or excessive data uploads. Additionally, the system provides visual insights through graphical representation of normal and attack requests, helping users understand system behaviour. This approach enhances cloud security by combining user interaction, monitoring intelligence, and preventive control measures in a simple and effective manner.