SuperInfer improves TTFT SLO attainment by up to 74.7% on GH200 Superchips via SLO-aware rotary scheduling (RotaSched) and full-duplex KV cache rotation (DuplexKV) over NVLink-C2C while preserving TBT and throughput.
V oltanallm: Feedback-driven frequency control and state-space rout- ing for energy-efficient llm serving.arXiv preprint arXiv:2509.04827
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
The energy cost of Large Language Model (LLM) inference is rapidly becoming a barrier to sustainable and scalable deployment. Although modern serving architectures expose distinct prefill and decode behaviors, existing systems fail to exploit these phase differences for energy-efficient serving under strict latency SLOs. This paper introduces VoltanaLLM, the first system that explicitly targets and reduces the energy bloat in modern prefill-decode (P/D) disaggregated LLM serving. Guided by a control-theory perspective, VoltanaLLM separates two levers: per-instance operating-point selection (GPU frequency per iteration) and system-level state-space routing of requests. We empirically observe that LLM inference exhibits a U-shaped energy-frequency curve creating "sweet spots" that depend on phase behavior and load. VoltanaLLM exploits this by combining phase-specific, iteration-level frequency selection driven by a lightweight, online-adaptive latency predictor, with a decode state-space guided router that avoids architectural granularity-induced inefficiencies, all while meeting desired SLOs. We implement VoltanaLLM using SGLang and evaluate it across multiple models and real-world workloads. Our results show VoltanaLLM reduces end-to-end energy by up to 36.3% versus a static max-frequency baseline while maintaining high SLO attainment, and generalizes to newer GPUs. These results point to sustainable LLM serving via phase-aware, iteration-level frequency selection coupled with architecture-aware routing. Source code is available in https://github.com/Supercomputing-System-AI-Lab/VoltanaLLM.
fields
cs.DC 4years
2026 4representative citing papers
A telemetry-driven cross-site router lets LLM inference run at wind-powered sites, cutting P99 latency by 22-52% over a strong baseline in a 64-GPU emulation.
KAIROS reduces power by 27% on average (up to 39.8%) for agentic AI inference by using long-lived context to jointly manage GPU frequency, concurrency, and request routing across instances.
Festina reduces energy consumption by up to 56% for serverless LLM inference on shared GPUs while keeping TTFT/TBT SLO attainment within 2% of four state-of-the-art baselines.
citing papers explorer
-
SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips
SuperInfer improves TTFT SLO attainment by up to 74.7% on GH200 Superchips via SLO-aware rotary scheduling (RotaSched) and full-duplex KV cache rotation (DuplexKV) over NVLink-C2C while preserving TBT and throughput.
-
CWind: A Cross-site Router for Large Language Model Inference Serving at Renewable Energy Farms
A telemetry-driven cross-site router lets LLM inference run at wind-powered sites, cutting P99 latency by 22-52% over a strong baseline in a 64-GPU emulation.
-
KAIROS: Stateful, Context-Aware Power-Efficient Agentic Inference Serving
KAIROS reduces power by 27% on average (up to 39.8%) for agentic AI inference by using long-lived context to jointly manage GPU frequency, concurrency, and request routing across instances.
-
Energy-Aware Scheduling for Serverless LLM Serving on Shared GPUs
Festina reduces energy consumption by up to 56% for serverless LLM inference on shared GPUs while keeping TTFT/TBT SLO attainment within 2% of four state-of-the-art baselines.