This technical report details the implementation of a high-performance computing environment utilizing a Kubernetes-orchestrated JupyterHub infrastructure. The platform constitutes a state-of-the-art research solution engineered to accommodate a broad spectrum of computational workloads, including quantum computing simulations and advanced artificial intelligence research. Deployed on a robust six-node cluster with specialized hardware procured by CNR-ICAR under the PNRR project "Humanities and Cultural Heritage Italian Open Science Cloud" (H2IOSC) this infrastructure supports Italy's strategic initiative to develop and integrate national Research Infrastructures (IR). The system seamlessly unifies traditional Jupyter notebooks with fully-featured web development environments, offering a cohesive platform for both scientific computing and software development. Its architecture strikes a deliberate balance among performance optimization, rigorous security enforcement, and user accessibility, making it equally suitable for computational researchers and developers working on resource-intensive applications.
Custom JupyterHub Deployment on H2IOSC Cluster
Francesco Gargiulo
;Luigi Barbato
2026
Abstract
This technical report details the implementation of a high-performance computing environment utilizing a Kubernetes-orchestrated JupyterHub infrastructure. The platform constitutes a state-of-the-art research solution engineered to accommodate a broad spectrum of computational workloads, including quantum computing simulations and advanced artificial intelligence research. Deployed on a robust six-node cluster with specialized hardware procured by CNR-ICAR under the PNRR project "Humanities and Cultural Heritage Italian Open Science Cloud" (H2IOSC) this infrastructure supports Italy's strategic initiative to develop and integrate national Research Infrastructures (IR). The system seamlessly unifies traditional Jupyter notebooks with fully-featured web development environments, offering a cohesive platform for both scientific computing and software development. Its architecture strikes a deliberate balance among performance optimization, rigorous security enforcement, and user accessibility, making it equally suitable for computational researchers and developers working on resource-intensive applications.| File | Dimensione | Formato | |
|---|---|---|---|
|
RT-ICAR-NA-JupyterHUB_EN.pdf
accesso aperto
Tipologia:
Versione Editoriale (PDF)
Licenza:
Dominio pubblico
Dimensione
460.04 kB
Formato
Adobe PDF
|
460.04 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


