Paper Detail

Energy-Aware Scheduling for Serverless LLM Serving on Shared GPUs

Tianyu Wang, Gourav Rattihalli, Aditya Dhakal, Longfei Shangguan, Dejan Milojicic

arxiv Score 9.0

Published 2026-06-29 · First seen 2026-06-30

Research Track A

Abstract

As LLM inference becomes a major cloud workload, its growing energy footprint makes cluster-wide energy optimization increasingly important. Serverless LLM serving helps platforms absorb traffic volatility by elastically sharing GPU resources across models, but this sharing also makes energy optimization difficult. Multiple co-resident models run under one device-wide operating point, while their resource demands and latency slack change across execution phases and load conditions. As a result, minimizing energy requires coordinated scheduling across request placement, runtime resource adaptation, and workload consolidation. We present Festina, a profiling-guided, power-aware control plane to minimize cluster-wide energy for serverless LLM serving. Unlike common global-local schedulers that focus on throughput or tail latency, Festina makes energy-first decisions by jointly coordinating request placement, SM partitioning, and GPU operating points under TTFT/TBT SLOs. In our system, a lightweight global scheduler performs fast, SLO-safe, energy-aware placement using constant-time lookups from offline profiles and GPU state summaries. On each GPU, a phase-aware local scheduler continuously adapts task batching and compute resources to minimize power consumption. Festina further performs energy-aware workload consolidation to reduce GPUs' static power consumption via SLO-aware migration. Comparison with four SOTA LLM serving systems and one DVFS-augmented system demonstrates that Festina reduces energy consumption by up to 56% while maintaining parity in SLO attainment (within a 2% margin)

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{wang2026energy,
  title = {Energy-Aware Scheduling for Serverless LLM Serving on Shared GPUs},
  author = {Tianyu Wang and Gourav Rattihalli and Aditya Dhakal and Longfei Shangguan and Dejan Milojicic},
  year = {2026},
  abstract = {As LLM inference becomes a major cloud workload, its growing energy footprint makes cluster-wide energy optimization increasingly important. Serverless LLM serving helps platforms absorb traffic volatility by elastically sharing GPU resources across models, but this sharing also makes energy optimization difficult. Multiple co-resident models run under one device-wide operating point, while their resource demands and latency slack change across execution phases and load conditions. As a result, },
  url = {https://arxiv.org/abs/2606.30391},
  keywords = {cs.DC},
  eprint = {2606.30391},
  archiveprefix = {arXiv},
}

Metadata

{}