logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-05072025-141206


Tipo di tesi
Tesi di laurea magistrale
Autore
BRETTONI, FRANCESCO
URN
etd-05072025-141206
Titolo
Testing Tools for Kubernetes Deployments: A Comparative Study
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
CYBERSECURITY
Relatori
relatore Prof. Brogi, Antonio
relatore Prof. Soldani, Jacopo
Parole chiave
  • Argo Workflows
  • cloud-native testing
  • comparative analysis
  • Frisbee
  • Kubernetes
  • test orchestration
  • Testkube
Data inizio appello
27/05/2025
Consultabilità
Completa
Riassunto
The widespread adoption of microservices and Kubernetes has revolutionized the development of cloud-native web applications, offering increased scalability, flexibility, and deployment automation. However, this architectural shift has introduced new testing challenges due to system complexity, diverse environments, and rapid deployment cycles. Traditional tools often lack the flexibility to orchestrate multiple test types in Kubernetes-based infrastructures. This work explores Kubernetes-native test orchestrators: tools designed to centrally manage and streamline diverse test scenarios in containerized environments. Three open-source solutions—Testkube, Argo Workflows, and Frisbee—were selected and evaluated through load, chaos, and security tests applied to two microservices applications. A dedicated taxonomy was developed to guide the comparative analysis, structured around categories such as orchestration, usability, performance, and security. The study found that each tool aligns with different use cases: Testkube is ideal for general-purpose testing with minimal setup, Argo offers deep customization for advanced teams, and Frisbee is suited for research and chaos experimentation. Beyond tool comparison, this work contributes to define and frame the concept of "test orchestrator" in Kubernetes contexts and provides a reusable evaluation framework. The findings provide guidance for practitioners and organizations selecting orchestration tools based on their testing requirements, technical environment, and organizational maturity.
File