March 2026

Conference Paper

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

By:
Bae, Jonghyun; Choi, Jong Youl ; Lupo Pasini, Massimiliano ; Mehta, Kshitij V; Ibrahim, Khaled
Page Number:
1193-1195
Book Title:
2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)
Publication Date:
March 12, 2026
Publisher Location:
IEEE, New Jersey, United States of America
Conference Name:
2024 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)
Conference Location:
San Francisco, California, United States of America
Conference Sponsor:
IEEE
View DOI Listing:
https://doi.org/10.1109/IPDPSW63119.2024.00203

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.