logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-01272026-192532


Tipo di tesi
Tesi di laurea magistrale
Autore
PIANIGIANI, GABRIELE
URN
etd-01272026-192532
Titolo
Testbed-as-a-Service: Automating Device Reconfiguration and Experiment Orchestration for Network Testbeds
Dipartimento
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
COMPUTER ENGINEERING
Relatori
relatore Prof. Mingozzi, Enzo
relatore Prof. Virdis, Antonio
relatore Ing. Lemmi, Laura
Parole chiave
  • Ansible automation
  • Ansible playbooks
  • automated provisioning
  • batch experiment execution
  • configuration rollbacks
  • configuration snapshots
  • experiment definition
  • experiment orchestration
  • network automation
  • network configuration
  • network experimentation
  • network testing platform
  • reproducible results
  • telemetry collection
  • testbed-as-a-service
  • time-based reservation
  • whitebox switches
Data inizio appello
27/02/2026
Consultabilità
Non consultabile
Data di rilascio
27/02/2029
Riassunto (Inglese)
Riassunto (Italiano)
Network experimentation requires dedicated infrastructure supporting reproducible, automated testing workflows. Traditional testbed deployment approaches are inadequate for the increasing complexity of network environments and rapid prototyping demands. This work presents a Testbed-as-a-Service platform that enables scalable and adaptive testbed provisioning through automated configuration of whitebox switches and hosts.
The system provides resource management, automated configuration, and batch experiment execution for physical network infrastructure. Users reserve the testbed for specific time intervals through a web-based interface, ensuring isolated access to devices. The architecture is microservices-based and includes independent components for authentication, scheduling, configuration, experiment orchestration, and telemetry collection, improving maintainability and scalability.
Reservations enable device configuration via restricted SSH CLI or via GUI-loaded Ansible playbooks, minimizing manual intervention, errors and optimizing deployment. Additionally, the system enables configuration snapshots and rollbacks.
Users define experiments through files including duration, playbooks, and execution timing, such files can be manually uploaded or created interactively with GUI. Batch execution sequentially runs experiments while collecting telemetry metrics. Results are presented as plots and command outputs, experiment definitions are bundled with results ensuring reproducibility.
File