Searcharxiv⌕ Search

arXiv subjects

M. Aiello

Publications and source records attributed to M. Aiello.

3 recordsLinked to original sources
arXiv2026

OpenGadget3 GPU solver tests

We present an in-depth evaluation of the scalability and accuracy of the GPU porting of the N-body code for hydrodynamic cosmological simulations \og. While technical details of our GPU porting were presented in Ragagnin et al. (2020), in this work we focus on assessing the accuracy of the ported modules: the short range gravity integrator, the different components of the hydrodynamic solver, and the conjugate gradient solver for thermal conduction. We ran several tests that gradually increase the number of physical modules included: a gravity-only cosmological simulation; a hydrodynamical shock tube test; a non-radiative zoom-in simulation of a galaxy cluster in a cosmological box; and a full-physics zoom-in simulation of a galaxy in a cosmological box. Comparing the results obtained with the GPU implementation to those from the classical CPU version, we find excellent agreement across all tests, with small differences on very small scales. For the individual physical modules, we find a GPU chip-to-chip speedup ranging from $\approx3-5$. For more complex cosmological and hydrodynamical setups, where a large number of physical processes and overheads contribute to the total workload, the observed total chip-to-chip speedup (with the same number of nodes and CPUs per node) is $\approx2-3$. We ran our tests on four different supercomputers: Leonardo Booster (CINECA), MareNostrum-V (BSC), SuperMUC-NG2 (LRZ), and the CIP cluster of the Faculty of Physics at the Ludwig-Maximilians-Universit\"at (LMU).

astro-ph.IM↗
arXiv2017

Personalized advice for enhancing well-being using automated impulse response analysis --- AIRA

The attention for personalized mental health care is thriving. Research data specific to the individual, such as time series sensor data or data from intensive longitudinal studies, is relevant from a research perspective, as analyses on these data can reveal the heterogeneity among the participants and provide more precise and individualized results than with group-based methods. However, using this data for self-management and to help the individual to improve his or her mental health has proven to be challenging. The present work describes a novel approach to automatically generate personalized advice for the improvement of the well-being of individuals by using time series data from intensive longitudinal studies: Automated Impulse Response Analysis (AIRA). AIRA analyzes vector autoregression models of well-being by generating impulse response functions. These impulse response functions are used in simulations to determine which variables in the model have the largest influence on the other variables and thus on the well-being of the participant. The effects found can be used to support self-management. We demonstrate the practical usefulness of AIRA by performing analysis on longitudinal self-reported data about psychological variables. To evaluate its effectiveness and efficacy, we ran its algorithms on two data sets ($N=4$ and $N=5$), and discuss the results. Furthermore, we compare AIRA's output to the results of a previously published study and show that the results are comparable. By automating Impulse Response Function Analysis, AIRA fulfills the need for accurate individualized models of health outcomes at a low resource cost with the potential for upscaling.

cs.CY↗