arXiv AI By Bole Ma, Jan Eitzinger, Harald Koestler, Gerhard Wellein

Diagnosing Overhead in Dispatch Operations: Cross-architecture Observatory

Read the original on arXiv AI →

arXiv:2605. 20982v2 Announce Type: replace-cross Abstract: AlltoAll dispatch is the dominant bottleneck of MoE expert parallelism, and the interconnect community has responded with four families of mitigations: predictive sample placement, adaptive expert relayout, hierarchical collectives, and EP-aware topology.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv Machine Learning
Jun 4

UltraEP: Unleash MoE Training and Inference on Rack-Scale Nodes with Near-Optimal Load Balancing

arXiv:2606. 04101v1 Announce Type: cross Abstract: Large-scale expert parallelism (EP) is becoming pivotal for training and serving frontier MoE models, but it also amplifies device-level expert load imbalance into compute stragglers, token all-to-all bottlenecks, and activation-memory spikes.

By Xinming Wei, Chao Jin, Tuo Dai, Yinmin Zhong, Shan Yu, Chengxu Yang, Bingyang Wu, Zili Zhang, Jing Mai, Qianchao Zhu, Zhouyang Li, Yuliang Liu, Guojie Luo