# MinatoLoader: Accelerating Machine Learning Training Through Efficient Data Preprocessing

arXiv: https://arxiv.org/abs/2509.10712 · HTML: https://mlsanity.com/paper/2509.10712/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 42 references that are in the map.
- First posted on arXiv 12 Sep 2025.

## Builds on (most cited first, 14 of 14)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation](https://mlsanity.com/paper/1606.06650/) (2016)
- [The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomes](https://mlsanity.com/paper/1904.00445/) (2019)
- [MLPerf Training Benchmark](https://mlsanity.com/paper/1910.01500/) (2019)
- [DataPerf: Benchmarks for Data-Centric AI Development](https://mlsanity.com/paper/2207.10062/) (2022)
- [Recurrent Neural Network Transducer for Audio-Visual Speech Recognition](https://mlsanity.com/paper/1911.04890/) (2019)
- [Chasing Carbon: The Elusive Environmental Footprint of Computing](https://mlsanity.com/paper/2011.02839/) (2020)
- [tf.data: A Machine Learning Data Processing Framework](https://mlsanity.com/paper/2101.12127/) (2021)
- [Analyzing and Mitigating Data Stalls in DNN Training](https://mlsanity.com/paper/2007.06775/) (2020)
- [Understanding Data Storage and Ingestion for Large-Scale Deep Recommendation Model Training](https://mlsanity.com/paper/2108.09373/) (2021)
- [RecD: Deduplication for End-to-End Deep Learning Recommendation Model Training Infrastructure](https://mlsanity.com/paper/2211.05239/) (2022)
- [Progressive Compressed Records: Taking a Byte out of Deep Learning Data](https://mlsanity.com/paper/1911.00472/) (2019)
- [nuts-flow/ml: data pre-processing for deep learning](https://mlsanity.com/paper/1708.06046/) (2017)
- [PreSto: An In-Storage Data Preprocessing System for Training Recommendation Models](https://mlsanity.com/paper/2406.14571/) (2024)

## Built on by (most cited first, 2 of 2)

- [A principled approach for energy-efficient training via phase-aware GPU frequency tuning](https://mlsanity.com/paper/2609.24205/) (2026)
- [Zephon: Elastic Determinism for Online, Stateful Foundation Model Data Loading Pipelines](https://mlsanity.com/paper/2610.03087/) (2026)

## Abstract (arXiv)

Data loaders are used by Machine Learning (ML) frameworks like PyTorch and TensorFlow to apply transformations to data before feeding it into the accelerator. This operation is called data preprocessing. Data preprocessing plays an important role in the ML training workflow because if it is inefficiently pipelined with the training, it can yield high GPU idleness, resulting in important training delays. Unfortunately, existing data loaders turn out to waste GPU resources, with 76% GPU idleness when using the PyTorch data loader, for example. One key source of inefficiency is the variability in preprocessing time across samples within the same dataset. Existing data loaders are oblivious to this variability, and they construct batches without any consideration of slow or fast samples. In this case, the entire batch is delayed by a single slow sample, stalling the training pipeline and resulting in head-of-line blocking. To address these inefficiencies, we present MinatoLoader, a general-purpose data loader for PyTorch that accelerates training and improves GPU utilization. MinatoLoader is designed for a single-server setup, containing multiple GPUs. It continuously prepares data in the background and actively constructs batches by prioritizing fast-to-preprocess samples, while slower samples are processed in parallel. We evaluate MinatoLoader on servers with V100 and A100 GPUs. On a machine with four A100 GPUs, MinatoLoader improves the training time of a wide range of workloads by up to 7.5× (3.6× on average) over PyTorch DataLoader and Pecan, and up to 3× (2.2× on average) over DALI. It also increases average GPU utilization from 46.4% with PyTorch to 90.45%, while preserving model accuracy and enabling faster convergence.