Saved in:
Bibliographic Details
Main Authors: Dai, Jun, Wang, Xiaorun, Fang, Kexiong, Yang, Zheng, Ji, Yuefeng, Zhang, Jiawei
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.12064
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918357009170432
author Dai, Jun
Wang, Xiaorun
Fang, Kexiong
Yang, Zheng
Ji, Yuefeng
Zhang, Jiawei
author_facet Dai, Jun
Wang, Xiaorun
Fang, Kexiong
Yang, Zheng
Ji, Yuefeng
Zhang, Jiawei
contents The proliferation of Large Language Models (LLMs) with exponentially growing parameters is making cross-data center (DC) training an inevitable trend. However, viable strategies for extending single-DC training frameworks to multi-DC environments remain underdeveloped. We experimentally demonstrate, for the first time, a high-performance geo-distributed LLMs training framework across multiple DCs interconnected by a lossless, remote direct memory access (RDMA) enabled Datacenter Optical Transport Network (DC-OTN). An enhanced pipeline parallelism scheme is implemented within the Ascend full-stack environment of Huawei, which effectively eliminates the impact of cross-DC communication overhead on training efficiency. The overlapped computation and cross-DC communication is achieved with constraint cross-DC bandwidth and High Bandwidth Memory (HBM), reducing computation bubble ratio by up to 78.91%.
format Preprint
id arxiv_https___arxiv_org_abs_2510_12064
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GeoPipe: a Geo-distributed LLM Training Framework with enhanced Pipeline Parallelism in a Lossless RDMA-enabled Datacenter Optical Transport Network
Dai, Jun
Wang, Xiaorun
Fang, Kexiong
Yang, Zheng
Ji, Yuefeng
Zhang, Jiawei
Networking and Internet Architecture
The proliferation of Large Language Models (LLMs) with exponentially growing parameters is making cross-data center (DC) training an inevitable trend. However, viable strategies for extending single-DC training frameworks to multi-DC environments remain underdeveloped. We experimentally demonstrate, for the first time, a high-performance geo-distributed LLMs training framework across multiple DCs interconnected by a lossless, remote direct memory access (RDMA) enabled Datacenter Optical Transport Network (DC-OTN). An enhanced pipeline parallelism scheme is implemented within the Ascend full-stack environment of Huawei, which effectively eliminates the impact of cross-DC communication overhead on training efficiency. The overlapped computation and cross-DC communication is achieved with constraint cross-DC bandwidth and High Bandwidth Memory (HBM), reducing computation bubble ratio by up to 78.91%.
title GeoPipe: a Geo-distributed LLM Training Framework with enhanced Pipeline Parallelism in a Lossless RDMA-enabled Datacenter Optical Transport Network
topic Networking and Internet Architecture
url https://arxiv.org/abs/2510.12064