Tropical: Enhancing SLO Attainment in Disaggregated LLM Serving via SLO-Aware Multiplexing

To guarantee service quality in transformer based large language model (LLM) serving, it is essential to meet the latency constraints of both the prefill phase (measured by Time-to-First-Token, TTFT) and the decode phase (measured by Time-per-Output-Token, TPOT). Non-disaggregated serving places pre...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	2025 62nd ACM/IEEE Design Automation Conference (DAC) S. 1 - 7
Hauptverfasser:	Ma, Jinming, Chen, Jiefei, Li, Xiuhong, Duan, Jiangfei, Duanmu, Haojie, Zhang, Xingcheng, Yang, Chao, Lin, Dahua
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 22.06.2025
Schlagworte:	Design automation Interference Large language models LLM Serving Multiplexing Phase measurement Queueing analysis Root cause analysis SLO-Aware Scheduling Transformers
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!