[2607.06202] UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods

Skip to main content

arXiv is now an independent nonprofit!<br>Learn more<br>&times;

Search arXiv

Press Enter to search &middot; Advanced search

-->

Computer Science > Distributed, Parallel, and Cluster Computing

arXiv:2607.06202 (cs)

[Submitted on 7 Jul 2026 (v1), last revised 8 Jul 2026 (this version, v2)]

Title:UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods

Authors:Yipeng Liu, Chang Liu, Si Shen, Jiaqi Zheng, Mingfan Li, Yuyang Yang, Guanhua Li, Yuquan Zhang, Yimeng Xu, Zhongzhe Hu, Zhiyuan Huang, Qihang Duan, Junsong Wang, Wenkai Ling, Baochuan Yang, Xianzhi Yu, Han Bao, Yijie Chen, Guihai Chen<br>View a PDF of the paper titled UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods, by Yipeng Liu and 18 other authors

View PDF

Abstract:The deployment of Mixture-of-Experts (MoE) models on production high-bandwidth superpods, such as NVIDIA's NVL72/576 and Huawei's CloudMatrix384, introduces critical challenges beyond raw interconnect bandwidth. While these systems provide unified global address spaces and high-bandwidth fabrics, their full potential for sparse MoE communication is hindered by three fundamental bottlenecks: (1) Strict execution serialization imposed by coarse-grained Bulk Synchronous Parallel (BSP) orchestration of interdependent communication phases; (2) Prohibitive synchronization overhead that fails to scale alongside high interconnect bandwidth; and (3) Severe load imbalance resulting from distance-agnostic scheduling of irregular token traffic. To eliminate these bottlenecks, we introduce UBEP (Unified-Bus Expert Parallelism), a production-ready communication library that rethinks MoE's All-to-All primitives for modern superpod architectures. Through large scale experiments, UBEP reduces All-to-All latency by up to 52.4% and MoE inference Time Per Output Token (TPOT) by up to 11.1%.

Comments:<br>Accepted to ACM SIGCOMM 2026. Corresponding authors: jzheng@nju.this http URL (J. Zheng), huzhongzhe@huawei.com (Z. Hu)

Subjects:

Distributed, Parallel, and Cluster Computing (cs.DC); Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI)

Cite as:<br>arXiv:2607.06202 [cs.DC]

(or<br>arXiv:2607.06202v2 [cs.DC] for this version)

https://doi.org/10.48550/arXiv.2607.06202

Focus to learn more

arXiv-issued DOI via DataCite

Related DOI:

https://doi.org/10.1145/3789240.3829183

Focus to learn more

DOI(s) linking to related resources

Submission history<br>From: Yipeng Liu [view email]<br>[v1]<br>Tue, 7 Jul 2026 12:25:16 UTC (762 KB)

[v2]<br>Wed, 8 Jul 2026 01:17:01 UTC (762 KB)

Full-text links:<br>Access Paper:

View a PDF of the paper titled UBEP: Re-architecting Expert Parallelism Communication Library for Production Superpods, by Yipeng Liu and 18 other authors<br>View PDF<br>TeX Source

view license

Current browse context:

cs.DC

next >

new<br>recent<br>| 2026-07

Change to browse by:

cs<br>cs.AI<br>cs.NI

References & Citations

NASA ADS<br>Google Scholar

Semantic Scholar

export BibTeX citation<br>Loading...

BibTeX formatted citation

&times;

loading...

Data provided by:

Bookmark

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Which authors of this paper are endorsers? |<br>Disable MathJax (What is MathJax?)

Major funding support from