Speaker
Description
The National Institute for Nuclear Physics (INFN) is a distributed research institute with computing resources spread across its federation. Through initiatives such as AI_INFN and projects like TeRABIT and ICSC, INFN has built a large-scale infrastructure for Machine Learning and AI, including GPU clusters, HPC systems, FPGA nodes, and high-performance storage. This infrastructure supports a growing community of experts applying ML to physics research, with a focus on scalability, open science, the exploration of emerging computing paradigms such as quantum computing and connecting experts with collaborations and dedicated hackathon events. By shifting toward innovative resource provisioning and advanced hardware R&D, AI_INFN ensures that researchers can effectively deploy machine learning algorithms, bridging the gap between cutting-edge technology and practical scientific applications.
Building on architectural advancements at the CNAF Tier-1 site, the project has optimized hardware orchestration and scalability while specifically enhancing the user experience for interactive and batch workloads. To validate this mature AI_INFN configuration, the ReCaS-Bari site—within the INFN Cloud federation—has been selected for platform replication and functional extension. This deployment serves as a critical testbed for frontier features, including workload offloading to the Leonardo supercomputer and other high-performance centers. By establishing ReCaS-Bari as a viable alternative to the CNAF instance, the project creates a foundation for future goals like cross-platform federation and the realization of a truly geo-distributed AI computing infrastructure for INFN.