 |
|||
|
|||
|
|||
Home > TERATEC FORUM > Workshops > Workshop 7
 |
TERATEC 2025 Forum |
|
Modern storage for IA Data Centers 2025-2030: a dominant architecture confronted by fragmented uses
Chaired by Jean-Thomas Acquaviva, Research engineer, DDN
and Farouk Mansouri, HPC-Cloud-AI senior engineer, LuxProvide
From Space Simulations to Deepfakes: Architecting AI Storage for a Fragmented Future
By Mohamed Adel Musallam, Research Fellow, Université du Luxembourg
AI systems —from synthetic satellite imagery to LLMs in classrooms and deepfake detection pipelines— place vastly different demands on storage infrastructure. In this talk, I present a researcher's perspective on how modern AI workloads interact with evolving storage environments. Drawing on practical use cases in computer vision, 3D CAD design, language models in education and finance, and simulation for space, I’ll explore how data is generated, consumed, and managed across domains. This presentation highlights the challenges, patterns, and lessons learned when deploying AI at scale in fragmented and demanding settings.
 |
Biography: Dr. Mohamed Adel is a Research Fellow in the Computer Vision, Imaging and Machine Intelligence Research Group (CVI²) at the Interdisciplinary Centre for Security, Reliability, and Trust (SnT), University of Luxembourg. His research lies at the intersection of computer vision, deep learning, and space technology. He has worked extensively on equivariant neural networks for improved geometric reasoning, synthetic data generation, and the development of robust AI models capable of reliable performance in extreme sensing environments typical of space missions. In addition to his contributions to space-related projects, Dr. Adel applies his AI expertise to broader domains, including anti-money laundering systems and intelligent educational platforms. |
|---|
Register now and get your badge here >>>