# Profiling and Improving the PyTorch Dataloader for high-latency Storage: A Technical Report

arXiv: https://arxiv.org/abs/2211.04908 · HTML: https://mlsanity.com/paper/2211.04908/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 9 Nov 2022.

## Builds on (most cited first, 9 of 9)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets from 3D Scans](https://mlsanity.com/paper/2110.04994/) (2021)
- [fastai: A Layered API for Deep Learning](https://mlsanity.com/paper/2002.04688/) (2020)
- [TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning](https://mlsanity.com/paper/2003.04696/) (2020)
- [Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Research](https://mlsanity.com/paper/1911.05063/) (2019)
- [High Performance I/O For Large Scale Deep Learning](https://mlsanity.com/paper/2001.01858/) (2020)
- [Accelerating Data Loading in Deep Neural Network Training](https://mlsanity.com/paper/1910.01196/) (2019)
- [Importance of Data Loading Pipeline in Training Deep Neural Networks](https://mlsanity.com/paper/2005.02130/) (2020)

## Built on by (most cited first, 5 of 5)

- [On Efficient Training of Large-Scale Deep Learning Models: A Literature Review](https://mlsanity.com/paper/2304.03589/) (2023)
- [I/O in Machine Learning Applications on HPC Systems: A 360-degree Survey](https://mlsanity.com/paper/2404.10386/) (2024)
- [The impact of the AI revolution on asset management](https://mlsanity.com/paper/2304.10212/) (2023)
- [Accelerating Recommender Model ETL with a Streaming FPGA-GPU Dataflow](https://mlsanity.com/paper/2501.12032/) (2025)
- [Optimizing Cloud-to-GPU Throughput for Deep Learning With Earth Observation Data](https://mlsanity.com/paper/2506.06235/) (2025)

## Abstract (arXiv)

A growing number of Machine Learning Frameworks recently made Deep Learning accessible to a wider audience of engineers, scientists, and practitioners, by allowing straightforward use of complex neural network architectures and algorithms. However, since deep learning is rapidly evolving, not only through theoretical advancements but also with respect to hardware and software engineering, ML frameworks often lose backward compatibility and introduce technical debt that can lead to bottlenecks and sub-optimal resource utilization. Moreover, the focus is in most cases not on deep learning engineering, but rather on new models and theoretical advancements. In this work, however, we focus on engineering, more specifically on the data loading pipeline in the PyTorch Framework. We designed a series of benchmarks that outline performance issues of certain steps in the data loading process. Our findings show that for classification tasks that involve loading many files, like images, the training wall-time can be significantly improved. With our new, modified ConcurrentDataloader we can reach improvements in GPU utilization and significantly reduce batch loading time, up to 12X. This allows for the use of the cloud-based, S3-like object storage for datasets, and have comparable training time as if datasets are stored on local drives.