Carolina Ferrari
LLM-based Intent-to-Policy Translation in Kubernetes Environments.
Rel. Alessio Sacco, Fulvio Valenza. Politecnico di Torino, Corso di laurea magistrale in Cybersecurity, 2026
|
Preview |
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview |
Abstract
Containerized applications have become a fundamental component of modern software systems, with Kubernetes serving as a standard platform for orchestrating them at scale. This widespread adoption has increased the need for effective network security policies that control communication between workloads within a cluster. Defining such policies, however, requires knowledge of low-level YAML formats that vary across Container Network Interface plugins, along with a precise understanding of the underlying network model. Most operators know what they want to enforce, few know exactly how to write it. This thesis addresses this gap by extending an existing tools, KubeTeus, with a multi-LLM pipeline for intent-based network policy management.
The proposed approach allows operators to express security requirements in natural language and automatically translates them into enforceable Kubernetes network policies
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Corso di laurea
Classe di laurea
URI
![]() |
Modifica (riservato agli operatori) |
