paper-with-me

Papers

Flash Communication: Reducing Tensor Parallelization Bottleneck for Fast Large Language Model Inference

2024-12-06 · Qingyuan Li, Bo Zhang, Liang Ye, Yifan Zhang, Wei Wu, Yerui Sun, Lin Ma, Yuchen Xie

The ever-increasing sizes of large language models necessitate distributed solutions for fast inference that exploit multi-dimensional parallelism, where computational loads are split across various accelerators such as GPU clusters. However, this approach often introduces significant communication overhead, especially on devices with limited bandwidth. In this paper, we introduce Flash Communication, a novel low-bit compression technique designed to alleviate the tensor-parallelism communication bottleneck during inference. Our method substantially boosts intra-node communication speed by more than 3x and reduces the time-to-first-token by 2x, with nearly no sacrifice in model accuracy. Extensive experiments on various up-to-date LLMs demonstrate the effectiveness of our approach.

📄 PDF Abstract BibTeX arXiv:2412.04964

Code (0)

등록된 구현이 없습니다.

Tasks

GPULanguage ModelingLanguage ModellingLarge Language Model

Methods 이 논문이 사용한 방법론

SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

CommFuse: Hiding Tail Latency via Communication Decomposition and Fusion for Distributed LLM Training

2026-04-27 · Rezaul Karim, Austin Wen, Wang Zongzuo, Weiwei Zhang 외 arxiv

The rapid growth in the size of large language models has necessitated the partitioning of computational workloads across accelerators such as GPUs, TPUs, and NPUs. However, these parallelization strategies incur substan…

Computational Efficiency

Unifying Data, Model and Hybrid Parallelism in Deep Learning via Tensor Tiling

2018-05-10 · Minjie Wang, Chien-chin Huang, Jinyang Li

Deep learning systems have become vital tools across many fields, but the increasing model sizes mean that training must be accelerated to maintain such systems' utility. Current systems like Tensorflow and MXNet focus o…

FlashCP: Load-Balanced Communication-Efficient Context Parallelism for LLM Training

2026-06-07 · Zheng Wang, Eric Liu, Linan Jiang, Zhongkai Yu 외 arxiv

Context parallelism (CP) is essential for training large-scale, long-context language models, as it partitions sequences to reduce memory overhead. However, existing CP methods suffer from workload imbalance, inefficient…

Speculative MoE: Communication Efficient Parallel MoE Inference with Speculative Token and Expert Pre-scheduling

2025-03-06 · Yan Li, Pengfei Zheng, Shuang Chen, Zewei Xu 외

MoE (Mixture of Experts) prevails as a neural architecture that can scale modern transformer-based LLMs (Large Language Models) to unprecedented scales. Nevertheless, large MoEs' great demands of computing power, memory …

Mixture-of-ExpertsScheduling

FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores

2023-11-10 · Daniel Y. Fu, Hermann Kumbong, Eric Nguyen, Christopher Ré

Convolution models with long filters have demonstrated state-of-the-art reasoning abilities in many long-sequence tasks but lag behind the most optimized Transformers in wall-clock time. A major bottleneck is the Fast Fo…