A PPO-based autoscaler for GPU inference in Kubernetes is claimed to cut P95 latency up to 6.7x, but the evidence is weakened by a missing HPA baseline, a spike-traffic slowdown, and a reliance on synthetic feedback.
Ai-enabling workloads on large-scale gpu-accelerated system: characterization, opportunities, and implications,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
KIS-S: A GPU-Aware Kubernetes Inference Simulator with RL-Based Auto-Scaling
A PPO-based autoscaler for GPU inference in Kubernetes is claimed to cut P95 latency up to 6.7x, but the evidence is weakened by a missing HPA baseline, a spike-traffic slowdown, and a reliance on synthetic feedback.