arXiv subjects

Caterina Caravita

Publications and source records attributed to Caterina Caravita.

4 recordsLinked to original sources

arXiv2025

EuroHPC SPACE CoE: Redesigning Scalable Parallel Astrophysical Codes for Exascale

High Performance Computing (HPC) based simulations are crucial in Astrophysics and Cosmology (A&C), helping scientists investigate and understand complex astrophysical phenomena. Taking advantage of exascale computing capabilities is essential for these efforts. However, the unprecedented architectural complexity of exascale systems impacts legacy codes. The SPACE Centre of Excellence (CoE) aims to re-engineer key astrophysical codes to tackle new computational challenges by adopting innovative programming paradigms and software (SW) solutions. SPACE brings together scientists, code developers, HPC experts, hardware (HW) manufacturers, and SW developers. This collaboration enhances exascale A&C applications, promoting the use of exascale and post-exascale computing capabilities. Additionally, SPACE addresses high-performance data analysis for the massive data outputs from exascale simulations and modern observations, using machine learning (ML) and visualisation tools. The project facilitates application deployment across platforms by focusing on code repositories and data sharing, integrating European astrophysical communities around exascale computing with standardised SW and data protocols.

astro-ph.IM↗

arXiv2025

Towards Exascale Computing for Astrophysical Simulation Leveraging the Leonardo EuroHPC System

Developing and redesigning astrophysical, cosmological, and space plasma numerical codes for existing and next-generation accelerators is critical for enabling large-scale simulations. To address these challenges, the SPACE Center of Excellence (SPACE-CoE) fosters collaboration between scientists, code developers, and high-performance computing experts to optimize applications for the exascale era. This paper presents our strategy and initial results on the Leonardo system at CINECA for three flagship codes, namely gPLUTO, OpenGadget3 and iPIC3D, using profiling tools to analyze performance on single and multiple nodes. Preliminary tests show all three codes scale efficiently, reaching 80% scalability up to 1,024 GPUs.

cs.DC↗

arXiv2022

A Parameter Space Exploration of High Resolution Numerically Evolved Early Type Galaxies Including AGN Feedback and Accurate Dynamical Treatment of Stellar Orbits

An extensive exploration of the model parameter space of axisymmetric Early-Type Galaxies (ETGs) hosting a central supermassive Black Hole (SMBH) is conducted by means of high resolution hydrodynamical simulations performed with our code MACER. Global properties such as 1) total SMBH accreted mass, 2) final X-ray luminosity and temperature of the X-ray emitting halos, 3) total amount of new stars formed from the cooling gas, 4) total ejected mass in form of supernovae and AGN feedback induced galactic winds, are obtained as a function of galaxy structure and internal dynamics. In addition to the galactic dark matter halo, the model galaxies are also embedded in a group/cluster dark matter halo; finally cosmological accretion is also included, with amount and time dependence derived from cosmological simulations. Angular momentum conservation leads to the formation of cold HI disks; these disks further evolve under the action of star formation induced by disk instabilities, of the associated mass discharge onto the central SMBH, and of the consequent AGN feedback. At the end of the simulations, the hot (metal enriched) gas mass is roughly $10\%$ the mass in the old stars, with twice as much having been ejected into the intergalactic medium. The cold gas disks are a $\approx$ kpc in size, and the metal rich new stars are in $0.1$ kpc disks. The masses of cold gas and new stars are roughly $0.1\%$ the mass of the old stars. Overall, the final systems appear to reproduce quite successfully the main global properties of real ETGs.

astro-ph.GA↗

arXiv2019

Dynamical models of spheroidal multi-component stellar systems

We present a significantly improved version of our numerical code JASMINE, that can now solve the Jeans equations for axisymmetric models of stellar systems, composed of an arbitrary number of stellar populations, a Dark Matter halo, and a central Black Hole. The stellar components can have different structural (density profile, flattening, mass, scale length), dynamical (rotational support, velocity dispersion anisotropy), and population (age, metallicity, Initial Mass Function, mass-to-light ratio) properties. These models, when combined with observations, will allow to investigate important issues, such as quantifying the systematic effects of IMF variations, of mass-to-light ratio gradients, and of different stellar kinematic components (e.g. counter rotating disks, kinematically decoupled cores) on luminosity-weighted properties. The developed analytical and numerical framework aims at modeling Early-Type Galaxies, but it can also be applied to dwarf Spheroidal galaxies and Globular Clusters.

astro-ph.GA↗