Xiaohongshu（小红书）

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

UltraEP：在机架级节点上以近最优负载均衡释放MoE训练与推理

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

机构 * School of Computer Science, Peking University（北京大学计算机学院） ； Xiaohongshu Inc.（小红书公司） ； Shanghai AI Laboratory（上海人工智能实验室） 纠错

AI总结 提出UltraEP，首个基于精确负载的实时均衡器，通过协同设计规划求解与专家复制通信，在机架级节点上实现MoE训练和推理的微批次与逐层重均衡，达到94.3%的力均衡理想吞吐量。

Comments The authors have identified issues related to information disclosure in the current version of the manuscript and therefore request its withdrawal. A revised version may be prepared at a later date