paper-with-me

Papers

Enabling Reconfiguration-Communication Overlap for Collective Communication in Optical Networks

2025-10-22 · Changbo Wu, Zhuolong Yu, Gongming Zhao, Hongli Xu arxiv

Collective communication (CC) is critical for scaling distributed machine learning (DML). The predictable traffic patterns of DML present a great opportunity for applying optical network technologies. Optical networks with reconfigurable topologies promise high bandwidth and low latency for collective communications. However, existing approaches face inherent limitations: static topologies are inefficient for dynamic communication patterns within CC algorithm, while frequent topology reconfiguration matching every step of the algorithm incurs significant overhead. In this paper, we propose SWOT, a demand-aware optical network framework that employs ``intra-collective reconfiguration'' to dynamically align network resources with CC traffic patterns. SWOT hides reconfiguration latency by overlapping it with data transmission through three key techniques: \textit{Heterogeneous Message Splitting}, \textit{Asynchronous Overlapping}, and \textit{Topology Bypassing}. Extensive simulations demonstrate that SWOT reduces communication completion time up to 89.7% across diverse CC algorithm compared to static baselines, demonstrating strong robustness to varying optical resources and reconfiguration delay.

📄 PDF Abstract BibTeX arXiv:2510.19322

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

FarSkip-Collective: Unhobbling Blocking Communication in Mixture of Experts Models

2025-11-14 · Yonatan Dukler, Guihong Li, Deval Shah, Jiang Liu 외 arxiv

Blocking communication presents a major hurdle in running MoEs efficiently in distributed settings. To address this, we present FarSkip-Collective which modifies the architecture of modern models to enable overlapping of…

mFabric: An Efficient and Scalable Fabric for Mixture-of-Experts Training

2025-01-07 · Xudong Liao, Yijun Sun, Han Tian, Xinchen Wan 외

Mixture-of-Expert (MoE) models outperform conventional models by selectively activating different subnets, named \emph{experts}, on a per-token basis. This gated computation generates dynamic communications that cannot b…

BlockingGPUMixture-of-Experts

ZeRO++: Extremely Efficient Collective Communication for Giant Model Training

2023-06-16 · Guanhua Wang, Heyang Qin, Sam Ade Jacobs, Connor Holmes 외

Zero Redundancy Optimizer (ZeRO) has been used to train a wide range of large language models on massive GPUs clusters due to its ease of use, efficiency, and good scalability. However, when training on low-bandwidth clu…

GPUQuantization

UCCL-Zip: Lossless Compression Supercharged GPU Communication

2026-04-19 · Shuang Ma, Chon Lam Lao, Zhiying Xu, Zhuang Wang 외 arxiv

The rapid growth of large language models (LLMs) has made GPU communication a critical bottleneck. While prior work reduces communication volume via quantization or lossy compression, these approaches introduce numerical…

T3: Transparent Tracking & Triggering for Fine-grained Overlap of Compute & Collectives

2024-01-30 · Suchita Pati, Shaizeen Aga, Mahzabeen Islam, Nuwan Jayasena 외

Large Language Models increasingly rely on distributed techniques for their training and inference. These techniques require communication across devices which can reduce scaling efficiency as the number of devices incre…