Quantify Co-Residency Risks in the Cloud Through Deep Learning

doi:10.1109/TDSC.2020.3032073

Quantify Co-Residency Risks in the Cloud Through Deep Learning

2021-07-01

9

OA

AI

DOI：10.1109/TDSC.2020.3032073

原文链接

查看原文

摘要

En 中文

Cloud computing, while becoming more and more popular as a dominant computing platform, introduces new security challenges. When virtual machines are deployed in a cloud environment, virtual machine placement strategies can significantly affect the overall security risks of the entire cloud. In recent years, the attacks are specifically designed to co-locate with target virtual machines in the cloud. The virtual machine placement without considering the security risks may put the users, or even the entire cloud, in danger. In this article, we present a fine-grained model to quantify the risk level caused by co-residency. Using a large scale dataset collected from Microsoft Azure Platform, we profile the behavior patterns of normal service subscribers (tenants) using our proposed feature metrics. Tenants are clustered into multiple categories. After the baseline is established based on the normal behavior pattern, the derivation can be evaluated for each category and the high-risk group can be labeled accordingly. With the labeled datasets, a classification component and a quantification component are constructed to dynamically quantify the co-residency risks for a specific virtual machine. Our experimental results demonstrate the robustness of our model to the new data and the accuracy is verified by examination of F-score Matrix.

Keyword:

Cloud computing

Security

Virtual machining

Measurement

Data mining

Deep learning

Computational modeling

Cloud security

deep learning

co-resident attack

Quantify Co-Residency Risks in the Cloud Through Deep Learning

摘要

AI总结

期刊

机构