TerraceMoE: A Cost Model for Hierarchical MoE All-to-All Communication

[2608.27874] TerraceMoE: A Cost Model for Hierarchical MoE All-to-All Communication Skip to main content Search Submit Donate Log in Computer Science > Distributed, Parallel, and Cluster Computing arXiv:2608.27874 (cs) [Submitted on 28 Aug 2026] Title: TerraceMoE: A Cost Model for Hierarchical MoE All-to-All Communication Authors: Weicheng Xue , Bingqiang Wang , Li Yuan , Huihui Zhou , Yonghong Tian View PDF HTML (experimental) Abstract: Hierarchical two-hop dispatch can reduce slow-fabric traffic in expert-parallel Mixture-of-Experts training, but it adds a second collective and an arrival-side operator chain. We present a cost model for screening that trade at the communication-call level, bounded by validation gates that withdraw a capability in code when they fail rather than reporting a caveat. At a reference geometry with 16 groups of 8 ranks, $q=3$, $H=2048$, and 4096 tokens per rank, the corrected effective breakeven hierarchy ratio is 3.98 for the measured PyTorch arrival chain, 1.49 for a hypothetical fused target, and 1.10 at zero implementation overhead. These are ratio-only sensitivity results, not deployment predictions: platform A measures 1.03, platform B has no separated fast/slow measurement, and neither machine measured here reaches the hierarchical regime. Four communication-level corpora pass their gates; a drift probe and the step-level gate fail. The latter failure is enforced in code, so we make no training-throughput prediction. The enabling routing constraint fixes per-token fan-out and per-selected-group quota, while aggregate per-peer counts remain data-dependent. Its measured validation-loss cost is small but nonzero (+0.0034 nats); downstream equivalence is reported with incomplete estimator provenance and is therefore not independently reconstructible from the artifact. Code, calibration constants and the validation gates are at this https URL . Comments: Preprint. Communication-level predictions only; step-level extrapolation is disabled by a failed validation gate. Fused-kernel figures are hypothetical sensitivity estimates. Code and executable gates released Subjects: Distributed, Parallel, and Cluster Computing (cs.DC) Cite as: arXiv:2608.27874 [cs.DC] (or arXiv:2608.27874v1 [cs.DC] for this version) https://doi.org/10.48550/arXiv.2608.27874 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Weicheng Xue [ view email ] [v1] Fri, 28 Aug 2026 03:27:27 UTC (160 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.DC < prev | next > new | recent | 2026-08 Change to browse by: cs References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation 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? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• TerraceMoE: A Cost Model for Hierarchical MoE All-to-All Communication#779783
• CascadeEP: Asynchronous Expert Execution for MoE Prefill under Attention Imbalance#1117964
• CascadeEP: Asynchronous Expert Execution for MoE Prefill under Attention Imbalance#1108732
• Post-Trained MoE Can Skip Half Experts via Self-Distillation#200493
• ELDR: Expert-Locality-Aware Decode Routing for PD-Disaggregated MoE Serving#329057