Attacks on Cloud and its Effects on IaaS

Abstract

In the era of Digitization and the e-commerce, all users are maintaining their data in a cloud and their personal information like documents, images, card information for quick payment and confidential data resides in the cloud. For Decades, most of the IT-industries, Government Agencies, Defense Sectors, Education hub, Hospitals and many more are using cloud services. One essential thing is that you require a network connection in order to get services from the cloud. There are many possibilities of attack in the cloud as the cloud is open and is connected to the internet. It means cloud run base on internet connection so chances of attacks on clouds are much more. Cloud is a hub of resources like network, storage, server, memory, CPU and any user uses these facilities worldwide at any time by subscription system. Cloud Computing provides so many services basically Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS) where the attacker always tries to breach the security at an IaaS level. Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks on IaaS which target resources like CPU, Memory, RAM usage and this is the major concern about the Cloud Service Provider (CSP). In this research at an initial level, Intrusion Detection System (IDS) technique is been deployed to detect the DoS and DDoS attack, for that Snort has been deployed. To overcome the problem of DoS and DDoS attack, Reinforcement Learning (RL) algorithm has been used to train the Cloud behavior for the classification of an attack. The basis on Reinforcement Learning Policy it shows either attack is Normal or High according to that it will gives Positive and Negative Reward. If an attack has been Normal then CSP go for Prevent that attack otherwise go for Load Balancing.