From Servers to Sites: Compositional Power Trace Generation of LLM Inference for Infrastructure Planning

Documenti analoghi

• Designing Datacenter Power Delivery Hierarchies for the AI Era
  di: Wilkins, Grant, et al.
  Pubblicazione: (2026)
• Offline Energy-Optimal LLM Serving: Workload-Based Energy Models for LLM Inference on Heterogeneous Systems
  di: Wilkins, Grant, et al.
  Pubblicazione: (2024)
• Energy Use of AI Inference: Efficiency Pathways and Test-Time Compute
  di: Oviedo, Felipe, et al.
  Pubblicazione: (2025)
• From Attention to Disaggregation: Tracing the Evolution of LLM Inference
  di: Kumar, Madabattula Rajesh, et al.
  Pubblicazione: (2025)
• Hybrid Heterogeneous Clusters Can Lower the Energy Consumption of LLM Inference Workloads
  di: Wilkins, Grant, et al.
  Pubblicazione: (2024)