A modified WAIT scheduler with online arrival-rate estimation matches or improves throughput over Sarathi-Serve, ORCA, and vLLM in low-shift bursty workloads.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
LLM Inference Under Bursty Workload Distribution: Modifying the WAIT Algorithm
A modified WAIT scheduler with online arrival-rate estimation matches or improves throughput over Sarathi-Serve, ORCA, and vLLM in low-shift bursty workloads.