paper-with-me

Papers

Training Time Prediction for Mixed Precision-based Distributed Training

2026-04-17 · Minchul Kang, Changyong Shin, Jinwoo Jeong, Hyunho Lee, Younghun Go, Gyeongmin Kim, Gyeongsik Yang, Chuck Yoo arxiv

Accurate prediction of training time in distributed deep learning is crucial for resource allocation, cost estimation, and job scheduling. We observe that the floating-point precision setting is a key determinant of training time, leading to training time variations of ~2.4x over its minimum. However, existing studies on distributed training time prediction rely on static model computation graphs that do not capture precision variations, including mixed precision. According to our experiments, training time prediction without considering precision results in significant prediction errors - reaching up to 147.85% in mean absolute percentage error (MAPE). To address this issue, we propose a precision-aware distributed training time predictor that achieves robust accuracy across diverse precision settings, including mixed precision, with 9.8% MAPE.

📄 PDF Abstract BibTeX arXiv:2604.16145

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Training Distributed Deep Recurrent Neural Networks with Mixed Precision on GPU Clusters

2019-11-30 · Alexey Svyatkovskiy, Julian Kates-Harbeck, William Tang

In this paper, we evaluate training of deep recurrent neural networks with half-precision floats. We implement a distributed, data-parallel, synchronous training algorithm by integrating TensorFlow and CUDA-aware MPI to …

GPUTime SeriesTime Series Analysis

Mixed-Precision Training for NLP and Speech Recognition with OpenSeq2Seq

2018-05-25 · Oleksii Kuchaiev, Boris Ginsburg, Igor Gitman, Vitaly Lavrukhin 외

We present OpenSeq2Seq - a TensorFlow-based toolkit for training sequence-to-sequence models that features distributed and mixed-precision training. Benchmarks on machine translation and speech recognition tasks show tha…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Machine Translationspeech-recognition+3

Distributed Low Precision Training Without Mixed Precision

2019-11-18 · Zehua Cheng, Weiyang Wang, Yan Pan, Thomas Lukasiewicz

Low precision training is one of the most popular strategies for deploying the deep model on limited hardware resources. Fixed point implementation of DCNs has the potential to alleviate complexities and facilitate poten…

GPUModel Compression

OpenSeq2Seq: Extensible Toolkit for Distributed and Mixed Precision Training of Sequence-to-Sequence Models

2018-07-01 · WS 2018 7 · Oleksii Kuchaiev, Boris Ginsburg, Igor Gitman, Vitaly Lavrukhin 외

We present OpenSeq2Seq {--} an open-source toolkit for training sequence-to-sequence models. The main goal of our toolkit is to allow researchers to most effectively explore different sequence-to-sequence architectures. …

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)DecoderMachine Translation+3

FP8-LM: Training FP8 Large Language Models

2023-10-27 · Houwen Peng, Kan Wu, Yixuan Wei, Guoshuai Zhao 외

In this paper, we explore FP8 low-bit data formats for efficient training of large language models (LLMs). Our key insight is that most variables, such as gradients and optimizer states, in LLM training can employ low-pr…

GPU