{"title":"Deep Learning-Based Vulnerability Detection and Mitigation in Virtualization Data Center","authors":"J Manikandan, U. Srilakshmi","doi":"10.5750/ijme.v1i1.1393","DOIUrl":null,"url":null,"abstract":"Virtualization is a critical technology that enables users to leverage the vast resources available within datacenters. Despite its numerous benefits, such as on-demand scalability, continuous availability, and cost efficiency, virtualization is susceptible to various security challenges, including intrusion, data compromise, and session hijacking. To address these threats, this study presents an innovative approach based on deep learning for detecting attacks and proactively isolating virtual machines (VMs) to mitigate their impact. The event sequences of VMs are transformed into event images using advanced techniques Integrated Gramian Markov Plot (IGMP). The proposed IGMP model comprises of the Gramian model with Markov estimate. The model uses the recurrence plot for the estimation of the IGMP in the virtualization process with the computation of data centers. Additionally, to improve the security IGMP model uses the aggregation signature generation model for the security features in the Virtual Machines. The proposed IGMP model uses the Deep learning models are then employed to extract meaningful features from these event images, which are subsequently classified into specific attack classes. Once an attack is predicted within the physical machine, the suspected VMs are immediately isolated to prevent further damage. Experimental results demonstrated that the high efficacy of the IGMP method, achieving an impressive attack prediction accuracy of 96%, surpassing existing approaches by at least 2%.","PeriodicalId":50313,"journal":{"name":"International Journal of Maritime Engineering","volume":null,"pages":null},"PeriodicalIF":0.7000,"publicationDate":"2024-07-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Maritime Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.5750/ijme.v1i1.1393","RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ENGINEERING, MARINE","Score":null,"Total":0}
引用次数: 0
Abstract
Virtualization is a critical technology that enables users to leverage the vast resources available within datacenters. Despite its numerous benefits, such as on-demand scalability, continuous availability, and cost efficiency, virtualization is susceptible to various security challenges, including intrusion, data compromise, and session hijacking. To address these threats, this study presents an innovative approach based on deep learning for detecting attacks and proactively isolating virtual machines (VMs) to mitigate their impact. The event sequences of VMs are transformed into event images using advanced techniques Integrated Gramian Markov Plot (IGMP). The proposed IGMP model comprises of the Gramian model with Markov estimate. The model uses the recurrence plot for the estimation of the IGMP in the virtualization process with the computation of data centers. Additionally, to improve the security IGMP model uses the aggregation signature generation model for the security features in the Virtual Machines. The proposed IGMP model uses the Deep learning models are then employed to extract meaningful features from these event images, which are subsequently classified into specific attack classes. Once an attack is predicted within the physical machine, the suspected VMs are immediately isolated to prevent further damage. Experimental results demonstrated that the high efficacy of the IGMP method, achieving an impressive attack prediction accuracy of 96%, surpassing existing approaches by at least 2%.
期刊介绍:
The International Journal of Maritime Engineering (IJME) provides a forum for the reporting and discussion on technical and scientific issues associated with the design and construction of commercial marine vessels . Contributions in the form of papers and notes, together with discussion on published papers are welcomed.