Viper: A High-Performance I/O Framework for Transparently Updating, Storing, and Transferring Deep Neural Network Models

J Ye, J Cernuda, N Rajesh, K Bateman… - Proceedings of the 53rd …, 2024 - dl.acm.org
Scientific workflows increasingly need to train a DNN model in real-time during an
experiment (eg using ground truth from a simulation), while using it at the same time for …

Scalable Access-Pattern Aware I/O Acceleration and Multi-Tiered Data Management for HPC and AI Workloads

A Maurya - 2024 - search.proquest.com
The exponential growth of data-intensive scientific simulations and deep learning workloads
presents significant challenges for high-performance computing (HPC) systems. These …