Improved t-SNE in Anomaly Detection of Cloud Virtual Machine

ABSTRACT

Aiming at high computation time in the process of cloud virtual machine anomaly detection, a C-t-SNE algorithm suitable for cloud is proposed. The algorithm classifies the virtual machines according to the different tasks, constructs the projection space according to the relevant parameters of the cloud virtual machines. Before dimensionality reduction, comparison datasets are extracted from each cloud virtual machine according to the relationship between the cloud virtual machine and the task. In the process of dimensionality reduction, all cloud virtual machines are replaced by comparison datasets, so as to reduce the number of data comparison and the calculation time.

KEYWORDS:

• Cloud computing
• virtual machine
• anomaly detection
• dimensionality reduction
• C-t-SNE

Acknowledgments

The authors would like to acknowledge the support by the Fundamental Research Funds for the Central Universities, Project number 2019GF07.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Fundamental Research Funds for the Central Universities [2019GF07].