2 papers
cs.PF2026
Simulate Before You Scale: High-Fidelity LLM Inference Planning Without GPUs
Dipanwita Guhathakurta, Mert Toslali, Srinivasan Parthasarathy +6
Deploying large language model (LLM) inference at scale requires jointly selecting hardware layout, parallelism strategy, scheduling policy, admission control, and request routing…
cs.AI2026
FluidPD: In-Place Elasticity for SLO-Aware Prefill-Decode Disaggregated LLM Serving
Kartik Ramesh, Kaidi Fu, Zihan Zheng +4
Prefill-decode disaggregation is becoming a common architecture for LLM serving because it separates two phases with distinct execution patterns and SLO objectives. Existing system…