AI 中文总结
研究针对MoE训练和服务中计划空间问题，提出可实现性感知全空间优化器moefs，通过三层搜索生成训练和服务栈，对开销定价，经两代硬件评估，其训练计划有提升且服务计划匹配，还给出了失败情况。
AI 中文摘要
混合专家（MoE）系统将程序的计划空间分为两部分：成本模型可以排序的空间和实际工具链能够构建的较小空间。自动优化器对第一部分进行排序，并默认两者一致，这可能导致返回理论上最优但无法实现的计划。我们提出了moefs，一种用于MoE训练和服务的可实现性感知全空间优化器，将部署可实现性作为首要搜索约束。moefs完成了对并行性、调度和内核的三层搜索；从单个计划中生成Megatron训练栈和SGLang服务栈；对测量的实现开销进行定价而非禁止。我们在两代硬件上进行评估。在2x RTX4090上，搜索到的训练计划比最强的手动调整基线高出0.9%；在8x H800上，搜索到的服务计划与手动调整配置的吞吐量比为1.0304。在8x H800训练中，搜索到的计划在最佳手动调整吞吐量的0.9338处可计算地失败。所有预测在H800运行前预先注册在冻结的、工件哈希裁决文件中，并如实报告每个结果。
英文摘要
Mixture-of-Experts (MoE) systems split a program's plan space in two: the space a cost model can rank, and the smaller space a real toolchain can actually build. Automatic optimizers rank the first and silently assume the two coincide -- so they can return a plan that is optimal on paper and impossible to emit. We present moefs, a realizability-aware full-space optimizer for MoE training and serving that makes deployment realizability a first-class search constraint. moefs closes a three-tier search over parallelism, schedule, and kernels; it emits both a Megatron training stack and an SGLang serving stack from a single plan; and it prices, rather than forbids, the realization overheads it measures. We evaluate across two hardware generations. On 2x RTX4090, the searched training plan edges the strongest hand-tuned baseline by +0.9% (and clears the 0.98x acceptance bar by +2.9%); on 8x H800, the searched serving plan matches the hand-tuned configuration at a 1.0304 throughput ratio. We hold failures to the same standard: on 8x H800 training, the searched plan is a computable, honest FAIL at 0.9338 of the best hand-tuned throughput, losing on a single schedule flag. All predictions are pre-registered in a frozen, artifact-hashed adjudication file before the H800 runs, and every outcome is reported as-is.
