Search NASA⌕ Search

DOE OSTI · 2428083

MDLoader: A Hybrid Model-driven Data Loader for Distributed Deep Neural Networks Training

Abstract

In this work, we propose MD Loader, a hybrid in-memory data loader for distributed deep neural networks. MDLoader introduces a model-driven performance estimator to automatically switch between one-sided and collective communication at runtime.

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Bae, Jonghyun, Choi, Jong Youl, Lupo Pasini, Massimiliano, Mehta, Kshitij, Ibrahim, Khaled. 2024-05-01. MDLoader: A Hybrid Model-driven Data Loader for Distributed Deep Neural Networks Training. https://doi.org/10.1109/ipdpsw63119.2024.00203

Cite the original work for its findings. Save a collection to share your selection of sources.