Reducing GPU Memory Fragmentation via Spatio-Temporal Planning for Efficient Large-Scale Model Training

• URL: http://arxiv.org/abs/2507.16274v1
• Date: Tue, 22 Jul 2025 06:39:07 GMT
• Title: Reducing GPU Memory Fragmentation via Spatio-Temporal Planning for Efficient Large-Scale Model Training
• Authors: Zixiao Huang, Junhao Hu, Hao Lin, Chunyang Zhu, Yueran Tang, Quanlu Zhang, Zhen Guo, Zhenhua Li, Shengen Yan, Zhenhua Zhu, Guohao Dai, Yu Wang,
• Abstract summary: We introduce STWeaver, a GPU memory allator for deep learning frameworks that reduces fragmentation by exploiting temporal regularity in memory allocation behaviors.<n>Built as a plug PyTorch, STWeaver reduces fragmentation ratio on average by 79.2% (up to 100%) across both dense and sparse models, with negligible overhead.
• Score: 9.775731832789116
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The rapid scaling of large language models (LLMs) has significantly increased GPU memory pressure, which is further aggravated by training optimization techniques such as virtual pipeline and recomputation that disrupt tensor lifespans and introduce considerable memory fragmentation. Default GPU memory allocators of popular deep learning frameworks like PyTorch use online strategies without knowledge of tensor lifespans, which can waste up to 43\% of memory and cause out-of-memory errors, rendering optimization techniques ineffective or even unusable. To address this, we introduce STWeaver, a GPU memory allocator for deep learning frameworks that reduces fragmentation by exploiting the spatial and temporal regularity in memory allocation behaviors of training workloads. STWeaver introduces a novel paradigm that combines offline planning with online allocation. The offline planning leverages spatio-temporal regularities to generate a near-optimal allocation plan, while the online allocation handles complex and dynamic models such as Mixture-of-Experts (MoE). Built as a pluggable PyTorch allocator, STWeaver reduces fragmentation ratio on average by 79.2\% (up to 100\%) across both dense and sparse models, with negligible overhead. This enables more efficient, high-throughput training configurations and improves performance by up to 32.5\%.

Related papers

• Dynamic Gradient Sparse Update for Edge Training [0.0502254944841629]
  gradient computation for backpropagation in the training requires significant memory buffers to store intermediate features and compute losses.<n>This is unacceptable for memory-constrained edge devices such as microcontrollers.<n>We propose a training acceleration method using dynamic gradient sparse updates to reduce memory usage.
  arXiv  Detail & Related papers  (2025-03-23T06:32:12Z)
• Memory-Efficient Training for Deep Speaker Embedding Learning in Speaker Verification [50.596077598766975]
  We explore a memory-efficient training strategy for deep speaker embedding learning in resource-constrained scenarios.<n>For activations, we design two types of reversible neural networks which eliminate the need to store intermediate activations.<n>For states, we introduce a dynamic quantization approach that replaces the original 32-bit floating-point values with a dynamic tree-based 8-bit data type.
  arXiv  Detail & Related papers  (2024-12-02T06:57:46Z)
• vTensor: Flexible Virtual Tensor Management for Efficient LLM Serving [53.972175896814505]
  Large Language Models (LLMs) are widely used across various domains, processing millions of daily requests. Large Language Models (LLMs) are widely used across various domains, processing millions of daily requests.
  arXiv  Detail & Related papers  (2024-07-22T14:37:58Z)
• SHERL: Synthesizing High Accuracy and Efficient Memory for Resource-Limited Transfer Learning [63.93193829913252]
  We propose an innovative METL strategy called SHERL for resource-limited scenarios. In the early route, intermediate outputs are consolidated via an anti-redundancy operation. In the late route, utilizing minimal late pre-trained layers could alleviate the peak demand on memory overhead.
  arXiv  Detail & Related papers  (2024-07-10T10:22:35Z)
• GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection [133.45193150403537]
  Training Large Language Models (LLMs) presents significant memory challenges due to the growing size of weights and GPU states. In this work, we propose Gradient Low-Rank Projection (GaLore) as a memory-efficient training strategy. Our 8-bit GaLore further reduces memory by up to 82.5% and total training memory by 63.3%, compared to a BF16 baseline.
  arXiv  Detail & Related papers  (2024-03-06T07:29:57Z)
• ROAM: memory-efficient large DNN training via optimized operator ordering and memory layout [8.99065455675796]
  We propose ROAM which operates on graph level to derive memory-efficient execution plan with optimized operator order and tensor memory layout for models. Experiments show that ROAM achieves a substantial memory reduction of 35.7%, 13.3%, and 27.2% compared to Pytorch and two state-of-the-art methods and offers a remarkable 53.7x speedup.
  arXiv  Detail & Related papers  (2023-10-30T06:29:21Z)
• Online Convolutional Re-parameterization [51.97831675242173]
  We present online convolutional re- parameterization (OREPA), a two-stage pipeline, aiming to reduce the huge training overhead by squeezing the complex training-time block into a single convolution. Compared with the state-of-the-art re-param models, OREPA is able to save the training-time memory cost by about 70% and accelerate the training speed by around 2x. We also conduct experiments on object detection and semantic segmentation and show consistent improvements on the downstream tasks.
  arXiv  Detail & Related papers  (2022-04-02T09:50:19Z)
• Mesa: A Memory-saving Training Framework for Transformers [58.78933015299703]
  We present Mesa, a memory-saving training framework for Transformers. Mesa uses exact activations during forward pass while storing a low-precision version of activations to reduce memory consumption during training. Experiments on ImageNet, CIFAR-100 and ADE20K demonstrate that Mesa can reduce half of the memory footprints during training.
  arXiv  Detail & Related papers  (2021-11-22T11:23:01Z)
• Layered gradient accumulation and modular pipeline parallelism: fast and efficient training of large language models [0.0]
  We analyse the shortest possible training time for different configurations of distributed training. We introduce two new methods, textitlayered gradient accumulation and textitmodular pipeline parallelism, which together cut the shortest training time by half.
  arXiv  Detail & Related papers  (2021-06-04T19:21:49Z)
• Revisiting Locally Supervised Learning: an Alternative to End-to-end Training [36.43515074019875]
  We propose an information propagation (InfoPro) loss, which encourages local modules to preserve as much useful information as possible. We show that InfoPro is capable of achieving competitive performance with less than 40% memory footprint compared to E2E training.
  arXiv  Detail & Related papers  (2021-01-26T15:02:18Z)
• Improving compute efficacy frontiers with SliceOut [31.864949424541344]
  We introduce SliceOut -- a dropout-inspired scheme to train deep learning models faster without impacting final test accuracy. At test time, turning off SliceOut performs an implicit ensembling across a linear number of architectures that preserves test accuracy. This leads to faster processing of large computational workloads overall, and significantly reduce the resulting energy consumption and CO2emissions.
  arXiv  Detail & Related papers  (2020-07-21T15:59:09Z)

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.