flowprefill-scheduling-preemption

【效率与协作】Improve LLM serving under mixed workloads by decoupling execution granularity from scheduling frequency. Operator-level preemption allows fine-grained interruption at natural boundaries (attention, feed-forward layers) without efficiency loss. Event-driven scheduling triggers decisions only on request arrival/completion. Eliminates head-of-line block…

独立 GitHub 快照 · 自动分类 · 以原作者文档为准