Factors Impacting I/O Time Proportion in AI Workloads

ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC), pp. 1–5

Abstract

The decision to optimize I/O in scientific applications often depends on the proportion of I/O time within an application's runtime. For AI workloads, such as machine learning (ML), we observe that different configurations with the same data size can lead to varying I/O impacts. In this work, we analyze common tunable parameters—batch size, number of samples, and number of files—typically adjusted when adapting ML models to new datasets or use cases. Using the XGBoost model, we predict the I/O percentage for ResNet50 and UNet3D workloads and identify feature combinations that influence I/O behavior. Our results provide practical guidance for optimizing I/O performance in AI workloads.

Authors 3

1. Zoya Masih

   University of Göttingen · Gesellschaft für wissenschaftliche Datenverarbeitung mbH Göttingen

   Affiliation as printed

   GWDG/ University of Goettingen, Göttingen, Germany

2. Radita Liem Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University, Aachen, Germany

3. Julian Martin Kunkel

   University of Göttingen · Gesellschaft für wissenschaftliche Datenverarbeitung mbH Göttingen

   Affiliation as printed

   GWDG / University of Goettingen, Göttingen, Germany