# DualScale: Energy-Efficient Disaggregated LLM Serving via Phase-Aware Placement and DVFS

arXiv: https://arxiv.org/abs/2602.18755 · HTML: https://mlsanity.com/paper/2602.18755/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Serving systems and request scheduling, one of 12 directions in Efficient LLM inference and serving.
- Builds on 14 of its 37 references that are in the map.
- First posted on arXiv 21 Feb 2026.

## Placed in

- Efficient LLM inference and serving / [Serving systems and request scheduling](https://mlsanity.com/inference/serving-systems-and-request-scheduling/)

## Builds on (most cited first, 14 of 14)

- [ShareGPT4V: Improving Large Multi-Modal Models with Better Captions](https://mlsanity.com/paper/2311.12793/) (2023)
- [Splitwise: Efficient generative LLM inference using phase splitting](https://mlsanity.com/paper/2311.18677/) (2023)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale](https://mlsanity.com/paper/2201.05596/) (2022)
- [PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU](https://mlsanity.com/paper/2312.12456/) (2023)
- [CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving](https://mlsanity.com/paper/2310.07240/) (2023)
- [LoongServe: Efficiently Serving Long-Context Large Language Models with Elastic Sequence Parallelism](https://mlsanity.com/paper/2404.09526/) (2024)
- [Accelerating Distributed MoE Training and Inference with Lina](https://mlsanity.com/paper/2210.17223/) (2022)
- [Parrot: Efficient Serving of LLM-based Applications with Semantic Variable](https://mlsanity.com/paper/2405.19888/) (2024)
- [Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training](https://mlsanity.com/paper/2208.06102/) (2022)
- [Towards Greener LLMs: Bringing Energy-Efficiency to the Forefront of LLM Inference](https://mlsanity.com/paper/2403.20306/) (2024)
- [Reducing Energy Bloat in Large Model Training](https://mlsanity.com/paper/2312.06902/) (2023)
- [SLO-aware GPU Frequency Scaling for Energy Efficient LLM Inference Serving](https://mlsanity.com/paper/2408.05235/) (2024)
- [Aladdin: Joint Placement and Scaling for SLO-Aware LLM Serving](https://mlsanity.com/paper/2405.06856/) (2024)

## Built on by (most cited first, 5 of 5)

- [Enabling Spatially Fine-Grained DVFS in Neural Processing Units for Energy-Efficient LLM Serving](https://mlsanity.com/paper/2607.16473/) (2026)
- [Energy-Efficient LLM Serving via Disaggregated Attention--FFN and Flexible Frequency Scaling](https://mlsanity.com/paper/2608.01891/) (2026)
- [Phase-Decoupled, Model-Calibrated Power Control for Disaggregated LLM Serving](https://mlsanity.com/paper/2609.11133/) (2026)
- [Analytical Power-Aware Provisioning for Prefill-Decode Disaggregated AI Inference](https://mlsanity.com/paper/2609.24639/) (2026)
- [Dissecting the Carbon-SLO Tension in Geo-Distributed LLM Serving](https://mlsanity.com/paper/2610.14571/) (2026)

## Abstract (arXiv)

Prefill/decode disaggregation is increasingly adopted in LLM serving to improve the latency-throughput tradeoff and meet strict TTFT and TPOT SLOs. However, LLM inference remains energy-hungry: autoscaling alone is too coarse-grained to track fast workload fluctuations, and applying fine-grained DVFS under disaggregation is complicated by phase-asymmetric dynamics and coupling between provisioning and frequency control. We present DualScale, a two-tier energy optimization framework for disaggregated LLM serving. DualScale jointly optimizes placement and DVFS across prefill and decode using predictive latency and power models. At coarse timescales, DualScale computes phase-aware placement and baseline frequencies that minimize energy while satisfying SLO constraints. At fine timescales, DualScale dynamically adapts GPU frequency per iteration using stage-specific control: model predictive control (MPC) for prefill to account for queue evolution and future TTFT impact, and lightweight slack-aware adaptation for decode to exploit its smoother, memory-bound dynamics. This hierarchical design enables coordinated control across timescales while preserving strict serving SLOs. Evaluation on a 16x H100 cluster serving Llama 3.3 70B with production-style traces shows that DualScale meets TTFT/TPOT SLOs while reducing energy by up to 39% in prefill and 48% in decode relative to DistServe.