EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Risk-Adaptive Zero-Trust Authorization for Dynamic Workloads Across Hybrid and Multi-Cloud Infrastructure

Shubh Prabhat

Sciences du Nord Nature Science and Technology · 31.12.2025 · 2 (02) · 195-207

DOI: 10.71238/snnst.v2i02.248 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sciences du Nord Nature Science and Technology, online erschienen am 31.12.2025. Verfasst von Shubh Prabhat. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional applications are short-lived, which run on containers across on-premises and multiple public cloud clusters. This movement cannot follow fixed rules and static positions of the network. Stolen workload credential could remain usable long and a breached container could access numerous of the neighbors without anyone noticing it. The paper introduces a dynamic zero trust authorization system of risk-adaptive workloads in the hybrid cloud and multi-cloud environment. The framework assigns marks on each request based on the four signals which include identity confidence, node posture, behavior anomaly and exposure context. It ranks the score on resource urgency, and time since a verification. A three-way decision is then used to permit or request step up verification or deny it and revoke the credential. Decision thresholds adjust to new traffic in such a way that the false denials during normal surges are not increased. We have scaled the framework on a testbed that comprised of one on-premises cluster and two public cloud clusters and have scaled 1,840 workload identities and 48.2 million authorization requests across 21 days. In five scenarios, there were 300 injected attacks that were in the test period. The framework scored an F1 score of 0.91 compared to 0.80 zero trust with fixed thresholds and 0.60 fixed role-based control. The mean time to revoke access dropped to 9.4 seconds and average blast radius dropped to 12 workloads. Staying low were the false denials of 0.8 percent, which remained low throughout the test period. The maximum delay added was 4.7 milliseconds per request. The outcomes indicate that the adaptive risk scoring is more effective in improving the containment with a small delay cost.

Zitieren

APA

Prabhat, S. (2025). Risk-Adaptive Zero-Trust Authorization for Dynamic Workloads Across Hybrid and Multi-Cloud Infrastructure. Sciences du Nord Nature Science and Technology, 2(02), 195-207. https://doi.org/10.71238/snnst.v2i02.248

Vancouver

Prabhat S. Risk-Adaptive Zero-Trust Authorization for Dynamic Workloads Across Hybrid and Multi-Cloud Infrastructure. Sciences du Nord Nature Science and Technology. 2025;2(02):195-207. doi:10.71238/snnst.v2i02.248

BibTeX
@article{Prabhat2025,
  author = {Prabhat, Shubh},
  title = {Risk-Adaptive Zero-Trust Authorization for Dynamic Workloads Across Hybrid and Multi-Cloud Infrastructure},
  journal = {Sciences du Nord Nature Science and Technology},
  year = {2025},
  volume = {2},
  number = {02},
  pages = {195--207},
  doi = {10.71238/snnst.v2i02.248}
}
Kostenlos starten →