VisIVO interactive high performance visualization

The exponential growth of data in Astrophysics and Cosmology demands scalable computational infrastructures coupled with interactive and reproducible visualization tools. This paper presents the integration of the VisIVO scientific visualization framework within the InterActive Computing (IAC) service at Cineca, providing real-time, GPU-accelerated visualization directly on HPC resources through a browser-based Jupyter interface. By developing dedicated Python wrappers and a custom Jupyter kernel, VisIVO can now be executed seamlessly within interactive notebooks, eliminating traditional command-line complexities and enabling immediate visualization on HPC compute nodes. Additionally, the incorporation of RESTful APIs based on the Flask framework enables the remote execution of VisIVO operations through lightweight web services, further abstracting backend complexity and enhancing interoperability with external applications. The framework improves user accessibility, fosters reproducibility, and supports high-throughput data exploration across large-scale astrophysical simulations. The system has been validated through representative use cases, including visual analysis of cosmological simulations produced with the GADGET code, demonstrating its scalability, robustness, and capacity to support interactive scientific discovery on HPC infrastructures.