Program last update: March 26th, 2026
Click here to download the program in PDF format
Day 1, Wednesday, March 25th
| 08:00 09:00 | Registration | |
| 09:00 09:30 | Opening ceremony – Aula Magna | |
| 09:30 10:30 | Keynote Talk: From algorithm to code to hardware to execution: why isn’t computing performance an experimental science? Prof. Gaétan Hains, University Paris-Est Créteil, France (room Aula Magna) |
|
| 10:30 11:20 | Coffee Break | |
| Main Track Session 1 (room Nicolae Iorga) Chair: Marco Danelutto |
Session SPDBNC (room Tiberiu Popoviciu) Chair: Igor Kotenko |
|
| 11:20 11:40 | HPC Meets Streaming: Benchmarking OpenMPI, Apache Flink, and Apache Storm, Eduardo Machado Martins, Breno Spohr Bernardi, Gabriel Rustick Fim, Renato B. Hoffmann, Gabriele Mencagli and Dalvan Griebler | Digital Twin of an Industrial Control System in the Power Grid: A Platform for Cybersecurity Threat Modeling and Assessment, Igor Kotenko, Igor Saenko and Evgenii Mityakov |
| 11:40 12:00 | In-Core, Low-Cost Hardware Transactional Memory, Álvaro Rubira-García, Eduardo José Gómez-Hernández, Rubén Titos-Gil and Alberto Ros | A Lightweight Hybrid Framework for Semantic Normalization of Security Events, Artem Mitelshtet and Sergei Savkov |
| 12:00 12:20 | ATTUNE: An Autotuning System for Energy–Performance Optimisation of Parallel Applications on Embedded and Mobile Platforms, Yazeed Almalaq and Vladimir Janjic | Application of Genetic Algorithms for Software Decompilation and their Improvement due to Large Language Models, Konstantin Izrailov and Igor Kotenko |
| 12:30 14:00 | Lunch Break – Piramida Restaurant | |
| Main Track Session 2 (room Nicolae Iorga) Chair: Dana Petcu |
Session SPDBNC (room Tiberiu Popoviciu) Chair: Igor Kotenko |
|
| 14:00 14:20 | Efficient Binary SpMV on GPUs via Binary Word Compression, Simone Staccone, Salvatore Filippone and Luca Pepè Sciarria | A Comparative Analysis of Small-Scale Language Models for Query Filtering in IoT Systems, Vladimir Lavrentiev and Dmitry Levshun |
| 14:20 14:40 | Research Data Version Management and Machine-Actionable Reproducibility for HPC, Andreas Knüpfer and Timothy J. Callow | A Methodology of Creating Digital Twins to Generate Realistic Datasets for Cybersecurity Incident Investigation, Maxim Gorda, Andrey Chechulin and Igor Kotenko |
| 14:40 15:00 | Reproducible Power Method for Computing Dominant Eigenvalue and Eigenvector, Mykhailo Havdiak, Oliver Lagerqvist, Stef Graillat and Roman Iakymchuk | Integrated Architecture for Testing AI Models in Cybersecurity, Evgenia Novikova, Mikhail Kuznetsov and Andrey Chechulin |
| 15:00 15:20 | Coffee Break | |
| Main Track Session 3 (room Nicolae Iorga) Chair: Gaétan Hains |
Session HPCMS (room Tiberiu Popoviciu) Chair: William Spataro |
|
| 15:20 15:40 | On Imperfect Byzantine Common Coins, Silvia Bonomi, Giovanni Farina, Roy Friedman, Eviatar Procaccia and Sebastien Tixeuil | Invited talk: MytilX: A Scalable HPC+AI Framework for Marine Bacterial Contamination Forecasting in Mussel Farming, Raffaele Montella, University of Naples “Parthenope” (UNP), Italy |
| 15:40 16:00 | On the Performance Evaluation of Dragonfly Connection Rules, Javier Navaridas and Jose A. Pascual | Transparent memory layout management for High Performance Computing, Andrea Giordano, Alessio De Rango, Thomas Garrafa and William Spataro |
| 16:00 16:20 | Multi-path Network Support for Unified Communication X (UCX), Tooraj Taraz, Elizabeth Reid and Ryan E. Grant | Accelerating the Particle-In-Cell code ECsim with OpenACC, Elisabetta Boella, Nitin Shukla, Filippo Spiga, Mozhgan Kabiri Chimeh, Matt Bettencourt and Maria Elena Innocenti |
| 16:20 16:40 | Optimizing the LiGen Drug Discovery Pipeline for Intel Max GPUs, Saleh Jamali Golzar, Lorenzo Carpentieri, Antonio De Caro, Biagio Cosenza, Davide Gadioli, Gianmarco Accordi, Gianluca Palermo, Federico Ficarelli, Daniele Gregori and Andrea R. Beccari | Performance and Energy-Aware Modeling of Sparse Matrix Multiplication in Social Graph Analysis, Elio Masciari and Enea Vincenzo Napolitano |
| 16:40 17:00 | Privacy-Preserving AI/GC Task Offloading in MEC via Generative Adversarial Differential Privacy, Wei Ai, Hongyi Chen, Yun Peng, Tao Meng, Zhixiong He and Keqin Li | A Visualization Module for Transparent Parallel Execution of Cellular Automata, Andrea Giordano, Alessio De Rango, Grzegorz Bazior, Davide Macri, Luigi Rizzo, Rocco Rongo, Jarosław Wąs, William Spataro and Giuseppe Mendicino |
| 17:00 17:20 | TimeGNN-Augmented Hybrid-Action MARL for Fine-Grained Task Partitioning and Energy-Aware Offloading in MEC, Wei Ai, Yun Peng, Tao Meng and Keqin Li | |
| 18:00 | Welcome Reception – Piramida Restaurant | |
Day 2, Thursday, March 26th
| 08:30 09:00 | Registration | ||
| 09:00 10:00 | Keynote Talk: Distribute and Learn 1 Billion Prof. Radu Prodan, University of Innsbruck, Austria (room Nicolae Iorga) |
||
| 10:00 10:30 | Coffee Break | ||
| Main Track Session 4 (room Nicolae Iorga) Chair: Alexandra Jimborean |
Session HPCMS (room Tiberiu Popoviciu) Chair: William Spataro |
||
| 10:30 10:50 | Evaluating Portable Programming Models for Hypergraph Label Propagation on GPUs, Antonio De Caro, Dario De Maio, Francesco Monzillo, Alessia Antelmi and Biagio Cosenza | Performance analysis of a parallel implementation of the city hostpot detector algorithm, Santina Capalbo, Eugenio Cesario, Paolo Lindia, Federica Lobello and Andrea Vinci | |
| 10:50 11:10 | Fast GPU-driven Kernel Launch with Dynamic Thread Allocation for Irregular Applications, Kento Kitamura, Kenji Tanaka and Kazunori Seno | A Multithreaded Implementation of the Analog Ensemble Method Assisted by a k-d Tree, Allan Rodrigues Clementino, Claudio Schepke, Carlos Balsa and José Carlos Rufino Amaro | |
| 11:10 11:30 | Infinite Data, Zero I/O: A CUDA-Optimized Framework for On-the-Fly Synthetic Data Generation for Deep Learning Pretraining, Ferran Soler-Guiral, Manuel F. Dolz and José I. Aliaga | Evaluating Deep Models for Motion-Based Authentication in Immersive Environments, Luca Ghiani, Enrico Grosso, Andrea Lagorio , Ruiu Pietro and Giuseppe A. Trunfio | |
| 11:30 11:50 | Proxy-Enabled Multi-Transport Connectivity, Nicolò Tonci and Massimo Torquati | Shard-Local Concurrent Hybrid k-d Tree, Kapil Kumar Attinagaramu, Praveen Alapati and Brahmaiah Gandham | |
| 12:00 13:30 | Lunch Break – Piramida Restaurant | ||
| 13:30 14:30 | Keynote Talk: Streaming Workflows: From Scientific Applications to AI and Back Prof. Marco Aldinucci, University of Turin, Italy (room Nicolae Iorga) |
||
| Session DT4DC (room Nicolae Iorga) Chair: Lavinia Chiara Tagliabue |
Session HPCMS (room Tiberiu Popoviciu) Chair: Andrea Giordano |
Session HPC@Edge (room D.V. Ionescu) Chair: Marco Lapegna |
|
| 14:40 15:00 | Real-Time BIM–IoT Synchronization for Operational Digital Twins: The HPC4AI Datacenter Case Study, Viviana Vaccaro, Lavinia Chiara Tagliabue and Robert Birke | Large-scale, mixed-precision brain simulations on heterogeneous accelerators, James Knight, Hengye Zhu and Thomas Nowotny | Exploring Distributed Energy Learning: A Federated, Self-Adapting Framework for Sensor-Free Power Estimation at Scale, Gennaro Mellone, Antonio Caccioppoli, Alberto Savarese, Fabio Angeletti, Pasquale Corvino and Diana Di Luccio |
| 15:00 15:20 | Sensor-Driven Digital Twin Architectures for Energy-Efficient Data Centers: The Use Case of EV Charging Infrastructure, Chiara Franzoni, Salvatore Dello Iacono and Stefano Rinaldi | Exploiting Warp-Level Parallelism for High-Order Finite Element Computations on GPUs, Balint Marton Nyakas, Dennis Giannacopoulos and Zeljko Zilic | Reinforcement Learning–Based Dynamic Management of Structured Parallel Farm Skeletons on Serverless Platforms, Lanpei Li, Massimo Coppola, Malio Li, Valerio Besozzi, Jack Bell and Vincenzo Lomonaco |
| 15:20 15:40 | Preliminary Life Cycle Assessment of a High-Performance Computing Data Center: A Case Study of the HPC4AI Facility at the University of Turin, Lavinia Chiara Tagliabue, Viviana Vaccaro, Alessandro Zichi, Robert Birke and Marco Aldinucci | Evaluating Performance in Transformer and Rule-Based Sentiment Analysis Models, Maria Chiara Martinis, Chiara Zucco and Mario Cannataro | Joint Resource-Aware Scheduling for Peer-to-Peer Stream Processing at the Edge, Mastaneh Bahmani and Julien Gascon-Samson |
| 15:40 16:00 | Coffee Break | ||
| Session DT4DC (room Nicolae Iorga) Chair: Lavinia Chiara Tagliabue |
Session HPCMS (room Tiberiu Popoviciu) Chair: William Spataro |
||
| 16:00 16:20 | LEED Certification and Data Centers: A Critical Assessment of Building Sustainability Rating Systems for HPC Facilities, Lavinia Chiara Tagliabue, Viviana Vaccaro, Alessandro Zichi and Silvia Meschini | Scalability and Energy Profiling of GNNs on Heterogeneous Hardware: Jetson vs x86, Luca Barillaro, Chiara Zucco, Marianna Milano, Giuseppe Agapito and Mario Cannataro | |
| 16:20 16:40 | Digital Twin for DCs: a reference architecture based on BIM, IoT and analytics control, Silvia Meschini, Paola Gasbarri, Lavinia Chiara Tagliabue, Viviana Vaccaaro, Robert Birke and Marco Aldinucci | From Molecules to Metrics: Parallel MO-Dijkstra for Multi-Objective Shortest Paths in Biology, Giuseppe Agapito, Mario Cannataro and Francesco Rania | |
| 16:40 17:00 | Simulation Model Tracking System: a Digital Twin-based technology to improve the thermal performance of Data Centres, Sergio Velásquez Correa, Rossano Scoccia, Marcello Aprile, Pablo Vicente Legazpi and Eduard Loscos Domingo | An Efficient and Scalable Federated Graph Neural Network Architecture for Information Operation Detection, Gianluigi Folino, Marco Minici, Pietro Sabatino and Michele Scinardo Tenghi | |
| 17:00 19:30 | City Tour (starting point: front of the UBB main building) | ||
| 19:30 | Conference Dinner Restaurant Panoramic Cetățuie |
||
Day 3, Friday, March 27th
| 08:30 09:00 | Registration | |
| 09:00 10:00 | Keynote Talk: Edge AI: research, technologies and applications Prof. Dana Petcu, West University of Timișoara, Romania (room Nicolae Iorga) |
|
| 10:00 10:40 | Sponsor talk: Marius Țurlea, Senior Solutions Architect, BRINEL | IQANTO, HPE & NVIDIA: HPE Private Cloud AI (room Nicolae Iorga) | |
| 10:40 11:00 | Coffee Break | |
| Main Track Session 5 (room Nicolae Iorga) Chair: Radu Prodan |
Session QC (room Tiberiu Popoviciu) Chair: Tudor Mihoc |
|
| 11:00 11:20 | Scalable Pipeline Parallel Training of Neural Networks using Reversibility, Ryota Miyagi, Kensuke Setsu, Kentaro Sano, Hiroshi Nakamura and Hideki Takase | Distributed Quantum–DNA Hybrid Architectures for Scalable Molecular Computing, Vasile Vancea |
| 11:20 11:40 | Accelerating Large-Scale Bayesian Network Structure Learning on GPU via Data Access Reduction, Akira Namekata, Ryota Miyagi, Hirotaka Nishikori, Hiroshi Nakamura and Hideki Takase | Scalable Quantum Inspired Evolutionary Algorithm for Multi-Objective Optimization Real World Problems, Olivian-Antonio Gilea, Adrian Florea, Claudia Banciu and Alexandru Telcean |
| 11:40 12:00 | Unlocking Parallel Efficiency in Monte Carlo Particle Transport Applications on Manycore Architectures, Mina Warnet, Adrien Roussel, Marc Perache and Michael Krajecki | Exploring Geometric Lattice-Inspired Distribute Quantum Computing: Analogies with Hydrophobic-Polar Models and Qubit Routing, Ioan Sima |
| 12:00 12:20 | Enabling AI Deep Potentials for Ab Initio-quality Molecular Dynamics Simulations in GROMACS, Andong Hu, Luca Pennati, Stefano Markidis and Ivy Peng | Quantum-Assisted Regression for Overall Survival Modeling in Acute Myeloid Leukemia, Tudor Mihoc and Rodica Ioana Lung |
| 12:30 14:00 | Lunch Break – Piramida Restaurant | |
| 14:00 14:30 | Awards and conclusions (Aula Magna) | |
Speakers for all sessions receive 15 minutes for their talk and 5 minutes for Q&A.
Sponsors:
 |
 |
 |
 |