{"ID":2874119,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.04827","arxiv_id":"2509.04827","title":"VoltanaLLM: Feedback-Driven Frequency Control and State-Space Routing for Energy-Efficient LLM Serving","abstract":"Modern Large Language Model (LLM) serving systems increasingly support interactive applications, like real-time chat assistants, code generation tools, and agentic workflows. However, the soaring energy cost of LLM inference presents a growing challenge for sustainable and cost-effective deployment. This paper introduces VoltanaLLM, a system for SLO-aware, energy-efficient LLM serving, built from a control theory perspective. VoltanaLLM co-designs frequency scaling and request routing in emerging prefill/decode disaggregated architectures, leveraging their decoupled execution to enable fine-grained phase-specific control. It consists of a feedback-driven frequency controller that dynamically adapts GPU frequency for prefill and decode phases, and a state-space router that explores routing decisions across frequency-scaled instances to minimize energy under latency constraints. We implement VoltanaLLM in SGLang and evaluate its performance over multiple state-of-the-art LLMs and real-world datasets. The results demonstrate that VoltanaLLM achieves up to 36.3% energy savings while maintaining near-perfect SLO attainment rate, paving the way for sustainable and intelligent LLM serving. Code of VoltanaLLM is open-sourced on GitHub: https://github.com/Supercomputing-System-AI-Lab/VoltanaLLM.","short_abstract":"Modern Large Language Model (LLM) serving systems increasingly support interactive applications, like real-time chat assistants, code generation tools, and agentic workflows. However, the soaring energy cost of LLM inference presents a growing challenge for sustainable and cost-effective deployment. This paper introduc...","url_abs":"https://arxiv.org/abs/2509.04827","url_pdf":"https://arxiv.org/pdf/2509.04827v2","authors":"[\"Jiahuan Yu\",\"Aryan Taneja\",\"Junfeng Lin\",\"Minjia Zhang\"]","published":"2025-09-05T05:58:16Z","proceeding":"cs.DC","tasks":"[\"cs.DC\",\"cs.AI\",\"cs.LG\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":610113,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2874119,"paper_url":"https://arxiv.org/abs/2509.04827","paper_title":"VoltanaLLM: Feedback-Driven Frequency Control and State-Space Routing for Energy-Efficient LLM Serving","repo_url":"https://github.com/Supercomputing-System-AI-Lab/VoltanaLLM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}