A Cloud-Enhanced Innovative Supercomputer System for the Integration of Simulation/Data/Learning/Inferencing
Tuesday, June 10, 2025 3:00 PM to Thursday, June 12, 2025 4:00 PM · 2 days 1 hr. (Europe/Berlin)
Foyer D-G - 2nd floor
Project Poster
AI Applications powered by HPC TechnologiesHeterogeneous System ArchitecturesHigh-Performance Data AnalyticsHPC in the Cloud and HPC ContainersHPC Simulations enhanced by Machine Learning
Information
Poster is on display.
Academic high-performance computing (HPC) infrastructures play a crucial role in conducting cutting-edge research on a diverse range of disciplinary fields including physics, chemistry, materials science, medical science, life science, atmospheric science, seismology, and social science. The exponential growth of data, the rapid advancement of machine learning, and the sophistication of computational models across these fields have made HPC increasingly complex yet indispensable for academia, industry, and even government, extending its impact far beyond the traditional domains.
This project aims to establish a new supercomputer at the University of Tokyo. We propose an integrated HPC system combining a GPU-based large-scale supercomputer with a container-based high-performance cloud system. This hybrid architecture enables us to flexibly handle various types of workloads such as scientific simulations, data analysis, machine learning training and inference, integration with AI-specific or quantum-based systems, and more complicated workloads that combine these, which meets the complex demands of the next-generating high-performance computing tasks.
Contributors:
Academic high-performance computing (HPC) infrastructures play a crucial role in conducting cutting-edge research on a diverse range of disciplinary fields including physics, chemistry, materials science, medical science, life science, atmospheric science, seismology, and social science. The exponential growth of data, the rapid advancement of machine learning, and the sophistication of computational models across these fields have made HPC increasingly complex yet indispensable for academia, industry, and even government, extending its impact far beyond the traditional domains.
This project aims to establish a new supercomputer at the University of Tokyo. We propose an integrated HPC system combining a GPU-based large-scale supercomputer with a container-based high-performance cloud system. This hybrid architecture enables us to flexibly handle various types of workloads such as scientific simulations, data analysis, machine learning training and inference, integration with AI-specific or quantum-based systems, and more complicated workloads that combine these, which meets the complex demands of the next-generating high-performance computing tasks.
Contributors:
Format
On DemandOn Site
