The paper develops fluid-guided online scheduling algorithms (WAIT and Nested WAIT) for LLM inference that handle endogenous KV-cache memory growth and improve stability and latency over baselines in simulations.
By Doob’s martingale inequality (Durrett 2019), we have that P( max 1≤i≤b−1 {eθkSi (k) } ≥a)≤ E M i a = E M0 a = 1 a
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Optimizing LLM Inference: Fluid-Guided Online Scheduling with Memory Constraints
The paper develops fluid-guided online scheduling algorithms (WAIT and Nested WAIT) for LLM inference that handle endogenous KV-cache memory growth and improve stability and latency over baselines in simulations.