{"id":330,"date":"2025-08-27T09:05:13","date_gmt":"2025-08-27T07:05:13","guid":{"rendered":"https:\/\/www.msnobile.eu\/?page_id=330"},"modified":"2026-02-26T11:01:19","modified_gmt":"2026-02-26T09:01:19","slug":"sdegno-stochastic-differential-equations-on-gpus-for-optimization","status":"publish","type":"page","link":"https:\/\/www.msnobile.eu\/index.php\/sdegno-stochastic-differential-equations-on-gpus-for-optimization\/","title":{"rendered":"SDEGnO – Stochastic Differential Equations on GPUs for Optimization"},"content":{"rendered":"\n
\"\"<\/figure>
\n

Optimization and performances testing of CUDA-(multi)GPU-accelerated codes for the automatic parameterization of physical models based on SDE Monte Carlo algorithms<\/p>\n<\/div><\/div>\n\n\n\n

The SDEGnO project<\/strong> aims to develop high-performance general-purpose simulators for the simulation and automatic calibration of physical models based on stochastic differential equations (SDEs)<\/strong>. The initiative seeks to implement an integrated framework that, through the use of Monte Carlo techniques<\/strong> and global optimization algorithms<\/strong> (including evolutionary strategies<\/strong> and swarm intelligence methodologies<\/strong>), enables a significant reduction in simulation time and energy consumption, while ensuring high accuracy in results.<\/p>\n\n\n\n

In the first phase<\/strong>, the team focused on optimizing CUDA code<\/strong>, fully leveraging the capabilities of NVIDIA multi-GPU architectures<\/strong>. This includes refining memory management and adopting Single Instruction, Multiple Data (SIMD)<\/strong> techniques to maximize computational efficiency. The next phase<\/strong> involves integrating advanced algorithms for the automatic search and calibration of physical parameters, allowing dynamic adaptation to various application scenarios. We will also develop new algorithms for the assessment of uncertainty <\/strong>and sensitivity <\/strong>of parameters.<\/p>\n\n\n\n

\"\"
The SDEGnO team gathered during a meeting in Venice on Feb 11, 2025. From bottom-left to bottom-right following a clockwise order:
Prof. Marco S. Nobile; Leone Bacciu; Dr. Giovanni Cavallotto; Prof. Sabina Rossi; Prof. Michele Bugliesi (guest); Dr. Stefano Dalla Torre; Matteo Grazioso. <\/figcaption><\/figure>\n\n\n\n

The resulting framework, replicable and scalable<\/strong>, will provide strategic support to future research groups in the HPC domain<\/strong> and contribute to strengthening the scientific competitiveness<\/strong> and economic impact<\/strong> of the sector. The main use case, motivating the work, is the prediction of cosmic radiation within the heliosphere<\/strong>, an investigation led by Dr. Stefano Della Torre of the INFN section of Milano.<\/p>\n\n\n\n

Bando a Cascata del Centro Nazionale ICSC – \u201cNational Centre for HPC, Big Data and Quantum Computing\u201d – codice CN00000013, Spoke 3 – \u201cAstrophysics and Cosmos Observation\u201d a valere sui fondi PNRR assegnati al Programma HPC finanziato sui fondi PNRR MUR \u2013 M4C2 \u2013 Investimento 1.4 con Decreto Direttoriale n. 1031 del 17\/06\/2022 – CUP C53C22000350006, pubblicato dall\u2019INAF \u2013 Istituto Nazionale di Astrofisica, approvato con Determina del Direttore Generale di INAF del 28 febbraio 2024, numero 31.<\/p>\n\n\n\n

Publications<\/h3>\n\n\n\n