 |
|||
|
|||
|
|||
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
Storage scaling and performance for massive-scale GPU infrastructure
By Elise Jennings, Head of HCP, NScale
Nscale, the hyperscaler engineered for AI, develops sustainable AI-ready data centres and delivers high-performance AI cloud services. This full-stack approach enables Nscale to co-design the data centre, GPU supercluster infrastructure and storage to deliver unrivalled speed, performance, and efficiency on AI workloads.
Nscale's full-stack AI private cloud platform is designed to support the entire generative AI lifecycle, enabling developers to run and train AI models. Performant data-centric workflows and AI workloads at scale demand intelligent management of infrastructure and storage. We discuss Nscale's approach for large-scale AI infrastructure and the importance of recent community efforts like MLCommons Storage to define standards, provide useful comparative metrics and track the latest innovation in software as well as hardware.
 |
Biography: Elise Jennings is a theoretical physicist and computer scientist who has worked at the forefront of High Performance Computing, AI and Quantum. Her work focuses on extreme-scale computing together with domain-driven AI to accelerate discovery across scientific disciplines. At the Leadership Computing Facility at Argonne National Laboratory and IBM Research Future of Computing, she led work on the integration and scaling of HPC simulation, AI software and Quantum computing on heterogeneous architectures in data-intensive science domains such as materials design, climate change, security, astronomy, high energy physics and biology. She is the Head of HPC at Nscale. |
|---|
Register now and get your badge here >>>