LayerScope: Predictive Cross-Layer Scheduling for Efficient Multi-Batch MoE Inference on Legacy Servers

Documenti analoghi

• Nezha: Breaking Multi-Rail Network Barriers for Distributed DNN Training
  di: Yu, Enda, et al.
  Pubblicazione: (2024)
• Alloc-MoE: Budget-Aware Expert Activation Allocation for Efficient Mixture-of-Experts Inference
  di: Liu, Baihui, et al.
  Pubblicazione: (2026)
• Multi-Layer Scheduling for MoE-Based LLM Reasoning
  di: Sun, Yifan, et al.
  Pubblicazione: (2026)
• EPS-MoE: Expert Pipeline Scheduler for Cost-Efficient MoE Inference
  di: Qian, Yulei, et al.
  Pubblicazione: (2024)
• From Tokens to Layers: Redefining Stall-Free Scheduling for MoE Serving with Layered Prefill
  di: Lee, Gunjun, et al.
  Pubblicazione: (2025)