Challenges, Advances and Sustainability in AI-HPC Interaction
In conjunction with the 25th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing — CCGrid 2025.
Call for Papers
The workshop name
Castanets are a percussion musical instrument that consists of two small round pieces of wood that you hold in the hand and hit together with the fingers to make a noise. In this workshop, the two shells represent AI and HPC interacting between them.
The CASTANET workshop focuses on the mutual interaction between High-Performance Computing (HPC) and Artificial Intelligence (AI), covering advances, best practices, and energy sustainability in these fields. The workshop brings together researchers to explore topics like energy-aware AI and HPC models, sustainable architectures, and resource optimization.
Topics of Interest
- AI-driven optimization of HPC resources
- HPC algorithms and software tools for AI and ML
- Mathematical methods for reducing dataset dimension
- Mixed-precision algorithms for AI and HPC
- Sustainable Machine Learning architectures
- AI for Energy Management
- HPC for big data analytics and visualization
- Performance and energy consumption trade-off
- Sustainable AI model training techniques
- Decentralized AI systems
- Intelligent resource allocation in sustainable computing
- Ethical considerations in AI and HPC energy consumption
- Energy-aware neural network design
- Interaction of AI and HPC in the Computing Continuum
- AI for climate change mitigation
Important Dates
- Paper submission deadline: February 24th (EXTENDED), 2025
- Acceptance notification: March 7th, 2025
- Camera-ready deadline: March 14th, 2025
Submission Guidelines
- Paper submission will follow the main conference rules.
- Papers must not exceed 10 pages, IEEE format, double-blind review.
- Submissions should be original and not under review elsewhere.
- Accepted papers will be published through IEEE Press.
Organizers
- Salvatore Cuomo — University of Naples Federico II, Italy
- Fabio Giampaolo — University of Naples Federico II, Italy
- Marco Lapegna — University of Naples Federico II, Italy
- Francesco Piccialli — University of Naples Federico II, Italy
Event Venue
CASTANET is co-located with the main conference IEEE International Symposium on Cluster, Cloud, and Internet Computing. Please refer to the CCGrid 2025 conference webpage for venue information and registration.
Program Committee
- Valeria Mele – University of Naples Federico II (Italy)
- David Camacho – Universidad Politecnica de Madrid (Spain)
- Jesus Carretero – Universidad Carlos III de Madrid (Spain)
- Diletta Chiaro – University of Naples Federico II (Italy)
- Jerry Chun-Wei Lin – Silesian University of Technology (Poland)
- Horacio González-Vélez – National College of Ireland (Ireland)
- Raffaele Montella – University of Naples Pathenope (Italy)
- Edoardo Prezioso – University of Naples Federico II (Italy)
- Giuliano Laccetti – University of Naples Federico II (Italy)
- Diego Romano – Italian National Research Council (Italy)
- Roman Wyrzykowski – Czestochowa University of Technology (Poland)
- Jiechen Zhao - University of Toronto (Canada)
Program
Monday 19th May — CAS1 (09:00–10:30), Chairman: Marco Lapegna
- 09:00 — Keynote (40 min): DAGonStar, a computational workflow engine designed for HPC and AI environmental modeling orchestration — Raffaele Montella (University of Naples Parthenope, Italy)
- 09:40 — Paper 1: DiasDNN-VD: Divide-and-conquer Model for Asynchronous Training of Large-Scale DNNs using Variational Dropouts — Sonali C. S., Ruchil Prajapati, Giri Prasad, Sathish Vadhiyar (Indian Institute of Science, India)
- 10:00 — Paper 2: Leveraging High-Performance Computing for Generating Large-Scale Synthetic Datasets of Focal Mechanisms in Seismic Networks — Daniela Annunziata, Edoardo Prezioso, Stefano Izzo, Marzia Canzaniello, Martina Savoia, Sara Amitrano, Pian Qi, Fabio Giampaolo, Francesco Piccialli (University of Naples Federico II, Italy)
CAS2 (11:00–12:30), Chairman: Francesco Piccialli
- 11:00 — Paper 3: Guidelines for the Quality Assessment of Energy-Aware NAS Benchmarks — Nick Kocher, Christian Wassermann, Leona Hennig, Jonas Seng, Holger Hoos, Kristian Kersting, Marius Lindauer, Matthias Müller
- 11:20 — Paper 4: Hierarchical Matrices in Graph Convolutional Deep Neural Network context: performance evaluation in a case study — Valeria Mele, Luisa Carracciuolo
- 11:40 — Paper 5: Enabling IoT Rejuvenation Through Machine Learning on Cloud/Edge Continuum: a Study to Fight the Proximity Sensor Ageing Installed in Intelligent Street Pole Lamps — Antonio Celesti, Giovanni Lonia, Antonino Quattrocchi, Roberto Montanini, Massimo Villari, Maria Fazio
- 12:00 — Paper 6: The prediction of bacteria contamination in farmed mussels at scale: HPC and AI join the forces — Ciro Giuseppe De Vita, Gennaro Mellone, Francisco Javier Garcia Blas
Contact Us
Marco Lapegna — marco.lapegna@unina.it