Kairos: SLO-Aware Scheduling and a Benchmark for Multi-Tenant LLM Inference Serving under Bursty, Heterogeneous Load

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Kairos: SLO-Aware Scheduling and a Benchmark for Multi-Tenant LLM Inference Serving under Bursty, Heterogeneous Load
Συγγραφείς: Gullapalli, Sneha, Divi, Veera Ravindra
Πηγή: 2026 5th International Conference on Electronics Representation and Algorithm (ICERA) Electronics Representation and Algorithm (ICERA), 2026 5th International Conference on. :598-603 Jul, 2026
Relation: 2026 5th International Conference on Electronics Representation and Algorithm (ICERA)
Βάση Δεδομένων: IEEE Xplore Digital Library
Περιγραφή
ISBN:9798319538901
9798319538918
DOI:10.1109/ICERA72709.2026.11666706