Cascade coordinates request scheduling and multi-tier KV-cache movement through a single per-request latency budget, improving SLO-satisfied goodput by up to 2.4x in simulation.
Llama 3 model card,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Cascade: Exploiting SLO-Aware latency budget for fair and high goodput LLM inference serving
Cascade coordinates request scheduling and multi-tier KV-cache movement through a single per-request latency budget, improving SLO-satisfied goodput by up to 2.4x in simulation.