MinatoLoader: Accelerating Machine Learning Training Through Efficient Data Preprocessing

• URL: http://arxiv.org/abs/2509.10712v2
• Date: Mon, 27 Oct 2025 21:57:37 GMT
• Title: MinatoLoader: Accelerating Machine Learning Training Through Efficient Data Preprocessing
• Authors: Rahma Nouaji, Stella Bitchebe, Ricardo Macedo, Oana Balmau,
• Abstract summary: MinatoLoader is a general-purpose data loader for PyTorch that accelerates training and improves GPU utilization.<n>It prioritizes fast-to-preprocessing samples, while slower samples are processed in parallel.<n>It improves the training time of a wide range of workloads by up to $7.5times$ ($3.6times$ on average) over PyTorch DataLoader and Pecan, and up to $3times$ ($2.2times$ on average) over DALI.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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 $7.5\times$ ($3.6\times$ on average) over PyTorch DataLoader and Pecan, and up to $3\times$ ($2.2\times$ 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.

Related papers

• Out of the Memory Barrier: A Highly Memory Efficient Training System for LLMs with Million-Token Contexts [68.79341332280062]
  Training Large Language Models (LLMs) on long contexts is severely constrained by prohibitive GPU memory overhead, not training time.<n>We introduce OOMB, a highly memory-efficient training system that directly confronts this barrier.<n>Our approach employs a chunk-recurrent training framework with on-the-fly activation recomputation, which maintains a constant activation memory footprint.
  arXiv  Detail & Related papers  (2026-02-02T13:52:40Z)
• TensorSocket: Shared Data Loading for Deep Learning Training [0.0]
  Deep learning training is a repetitive and resource-intensive process.<n>In this paper, we presentSocket to reduce the computational needs of training by enabling simultaneous training processes to share the same data loader.<n>Our evaluation shows that colSocket enables scenarios that are infeasible without data sharing, increases training throughput by up to 100%, and when utilizing cloud instances, achieves cost savings of 50%.
  arXiv  Detail & Related papers  (2024-09-27T13:39:47Z)
• Efficient Tabular Data Preprocessing of ML Pipelines [9.23424733090734]
  Data preprocessing pipelines are a crucial component of Machine Learning (ML) training. Piper is a hardware accelerator for data preprocessing, prototype it on FPGAs, and demonstrate its potential for training pipelines of commercial recommender systems. Piper achieves 4.7 $sim$ 71.3$times$ speedup in latency over a 128-core CPU server and outperforms a data-center GPU by 4.8$sim$ 20.3$times$ when using binary input.
  arXiv  Detail & Related papers  (2024-09-23T11:07:57Z)
• Efficient Asynchronous Federated Learning with Sparsification and Quantization [55.6801207905772]
  Federated Learning (FL) is attracting more and more attention to collaboratively train a machine learning model without transferring raw data. FL generally exploits a parameter server and a large number of edge devices during the whole process of the model training. We propose TEASQ-Fed to exploit edge devices to asynchronously participate in the training process by actively applying for tasks.
  arXiv  Detail & Related papers  (2023-12-23T07:47:07Z)
• FFCV: Accelerating Training by Removing Data Bottlenecks [84.89623507733963]
  We present FFCV, a library for easy and fast machine learning model training. It speeds up model training by eliminating (often subtle) data bottlenecks from the training process. Detailed installation instructions, documentation, and Slack support channel are available at https://ffcv.io/.
  arXiv  Detail & Related papers  (2023-06-21T19:06:41Z)
• tf.data service: A Case for Disaggregating ML Input Data Processing [4.851146762916078]
  Machine learning (ML) computations commonly execute on expensive specialized hardware, such as GPUs and TPUs, which provide high FLOPs and performance-per-watt. To avoid data stalls, the host CPU and RAM required for input data processing per accelerator core used for ML computations varies across jobs. We present tf.data service, an open-source disaggregated input data processing service built on top of tf.data in.
  arXiv  Detail & Related papers  (2022-10-26T16:15:45Z)
• PARTIME: Scalable and Parallel Processing Over Time with Deep Neural Networks [68.96484488899901]
  We present PARTIME, a library designed to speed up neural networks whenever data is continuously streamed over time. PARTIME starts processing each data sample at the time in which it becomes available from the stream. Experiments are performed in order to empirically compare PARTIME with classic non-parallel neural computations in online learning.
  arXiv  Detail & Related papers  (2022-10-17T14:49:14Z)
• Where Is My Training Bottleneck? Hidden Trade-Offs in Deep Learning Preprocessing Pipelines [77.45213180689952]
  Preprocessing pipelines in deep learning aim to provide sufficient data throughput to keep the training processes busy. We introduce a new perspective on efficiently preparing datasets for end-to-end deep learning pipelines. We obtain an increased throughput of 3x to 13x compared to an untuned system.
  arXiv  Detail & Related papers  (2022-02-17T14:31:58Z)
• Importance of Data Loading Pipeline in Training Deep Neural Networks [2.127049691404299]
  In large models, the time spent loading data takes a significant portion of model training time. We compare binary data format to accelerate data reading, and NVIDIA DALI to accelerate data augmentation. Our study shows improvement on the order of 20% to 40% if such dedicated tools are used.
  arXiv  Detail & Related papers  (2020-04-21T14:19:48Z)
• MOGPTK: The Multi-Output Gaussian Process Toolkit [71.08576457371433]
  We present MOGPTK, a Python package for multi-channel data modelling using Gaussian processes (GP) The aim of this toolkit is to make multi-output GP (MOGP) models accessible to researchers, data scientists, and practitioners alike.
  arXiv  Detail & Related papers  (2020-02-09T23:34:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.