Data analysis of cloud virtualization experiments and network latency

Pedro R. X. do Carmo and co-authors submitted an arXiv paper (arXiv:2602.05792) presenting a dataset and analysis of cloud virtualization experiments on 5 Feb 2026.

• Main announcement: The authors published a dataset of active network measurements and an accompanying analysis on arXiv (arXiv:2602.05792, submitted 5 Feb 2026); the dataset varies CPU affinity, echo packet injection frequency, virtual network driver type, CPU/I/O/network load, and number of concurrent VMs, and evaluates end-to-end latency / round-trip time. Access: PDF at https://arxiv.org/pdf/2602.05792, HTML at https://arxiv.org/html/2602.05792v1, TeX Source at https://arxiv.org/src/2602.05792; license: CC BY-NC-SA 4.0; DOI via DataCite pending: https://doi.org/10.48550/arXiv.2602.05792.
• Background and details: The study uses virtualization technologies KVM, LXC, and Docker, applies pre-processing, correlation analysis, dimensionality reduction, and clustering, and provides a brief analysis intended to support developing machine learning-based systems for administrator decision-making. The paper is categorized under Networking and Internet Architecture (cs.NI); Databases (cs.DB).