Cutting AI Datacenter Energy with Reinforcement Learning: Measured Power Control of LLM Training from One GPU to the Fleet

• URL: http://arxiv.org/abs/2608.11226v1
• Date: Mon, 27 Jul 2026 20:51:03 GMT
• Title: Cutting AI Datacenter Energy with Reinforcement Learning: Measured Power Control of LLM Training from One GPU to the Fleet
• Abstract summary: We instrument GRPO training with half-second power telemetry at 7B, 14B, and 72B scales on one to four A100s (380,000+ samples)<n>We train a PPO meta-controller that adapts the workload's own generation parameters to measured power.<n>We quantify the economic and carbon consequences and specify a low-cost operator pilot.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Reinforcement-learning post-training dominates modern language-model development, yet its power behavior on GPU hardware has not been characterized, and datacenters manage GPU power with workload-blind mechanisms, static caps and reactive throttling, that slow hardware indiscriminately. We instrument GRPO training with half-second power telemetry at 7B, 14B, and 72B scales on one to four A100s (380,000+ samples), and train a PPO meta-controller that adapts the workload's own generation parameters to measured power. Against the full 500-step 7B trace, the controller cuts power-limit violations by 89.8% while increasing token output by 18.1% and energy efficiency by 26.2% (tokens per MWh). Deployed live at 72B, the same controller family yields replicated null results, diagnosed as the group-size actuator losing authority under model sharding. An actuator-authority sweep shows the same parameters applied as generation concurrency retain 17-22% power authority, isolating an occupancy-versus-volume principle; a controller rebuilt on that actuator controls a live 72B rollout-generation workload across three replications: 35.7% more output than a static safe baseline at 2.27 +/- 1.08% budget violations, 87.2% fewer violations than uncontrolled operation, and the best mean throughput and energy per token among constrained controllers, with an adaptive threshold rule matching it in one of three operating conditions. Under realistic measurement windows the original 72B transients fall from 23.6% at half-second resolution to 1.6% at 30 s and zero at 5 min; a composed 16-GPU fleet shows zero violations at 30 s and longer, with peak demand at 50-56% of nameplate. For this fleet mix, roughly twofold oversubscription of nameplate appears feasible, subject to operator validation. We quantify the economic and carbon consequences and specify a low-cost operator pilot.

Related papers

• nBMS, a Neuromorphic Battery Management System with a Silicon-Validated Spiking State-of-Charge Core for eVTOL Aircraft [0.0]
  Event-driven spiking network is designed for per-timestep state-of-charge estimation.<n>System is deployed on a low-cost automotive-qualified Artix-7 field-programmable gate array.<n>Fully annotated post-implementation analysis gives 1.30$$J per inference, an average of 0.65$$W at the 0.5Hz mission cadence.
  arXiv  Detail & Related papers  (2026-09-11T20:12:16Z)
• Phase-Decoupled, Model-Calibrated Power Control for Disaggregated LLM Serving [1.249668212599889]
  We deploy NVIDIA's Max-Q inference profile on a disaggregated B200 system.<n>On an 8x B200 node serving Qwen3-Coder-480B (FP8) under agentic load, our balanced mode delivers +20.4% tokens/J at +3.5% mean e2e.
  arXiv  Detail & Related papers  (2026-09-10T06:23:37Z)
• Multi-Domain Physics-Based MDO of Multirotor UAVs: A Deterministic Framework for Discrete COTS Sizing [1.6114012813668932]
  This paper presents AeroEval, a physics-based, Multidisciplinary Design Optimization (MDO) engine.<n>Engine maps continuous sizing optima to physically purchasable, commercial off-the-shelf (COTS) components.<n>Battery mass is predicted within 7.9% on single-pack platforms.
  arXiv  Detail & Related papers  (2026-07-24T03:52:36Z)
• Mach-Mind-4-Flash Technical Report [0.0]
  Mach-Mind-4-Flash is a Mixture-of-Experts (MoE) agentic model with 3B activated parameters.<n>Post-training optimization alone without scaling pre-training compute, the model achieves performance on par with or surpassing that of 100B- parameter-class models.
  arXiv  Detail & Related papers  (2026-07-10T12:57:08Z)
• LocalNav: Distilling Frontier VLMs and Embodied RL for On-Device Object Goal Navigation [53.88286427894783]
  Vision Language Models (VLMs) have emerged in the robotic domain as a powerful tool that enables environmental perception with language context.<n>Yet, their computational footprint typically confines them to cloud execution, hindering low-latency inference with local deployment on resource-constrained robots.<n>We present a distillation strategy that transfers complex spatial-semantic reasoning from large frontier models into a lightweight, 4B- parameter local VLM for edge execution on embedded GPU devices.
  arXiv  Detail & Related papers  (2026-06-26T09:11:59Z)
• Reliability-Asymmetric Spacecraft Autonomy: Co-Designing a Capable Learned GNC Stack with a Verified, Adaptation-Aware Runtime Shield [2.28438857884398]
  Rule-based autonomy is certifiable but brittle, while learned autonomy is capable but hard to verify.<n>We present AMPLE-GNC, a three-tier guidance, navigation, and control stack.<n>Fine-tuning a pretrained 360M model with grammar-constrained decoding gives a hard output-validity guarantee.<n>We show that a latching safe-hold shield can suppress even a capable controller.
  arXiv  Detail & Related papers  (2026-06-24T03:55:21Z)
• ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models [45.62029693245481]
  We introduce ActQuant, an action-guided mixed-precision PTQ framework.<n>ActQuant operates in two stages: (1) an inter-tensor bit allocator that assigns each weight matrix a single bitwidth based on how much it contributes to predicting the agent's actions.<n>We also introduce OmniModel, an agentic conversion pipeline that ports into a native C/C++ with efficient low-bit kernels.
  arXiv  Detail & Related papers  (2026-05-19T19:57:26Z)
• Brain-inspired spike-timing plasticity for reliable label-efficient event-camera vision [0.3823356975862005]
  Local spike-timing-dependent plasticity (STDP) modules operate on a single CPU thread without GPU support.<n>Dense gradient-trained detectors cannot provide this combination of gradient training, dense matrix multiplication, and local plasticity-free operation by construction.
  arXiv  Detail & Related papers  (2026-05-17T22:56:16Z)
• A Delta-Aware Orchestration Framework for Scalable Multi-Agent Edge Computing [0.0]
  The Synergistic Collapse occurs when scaling beyond 100 agents causes superlinear performance degradation that individual optimizations cannot prevent.<n>Prior work has addressed each contributing factor in isolation: exponential action-space growth, computational redundancy among spatially adjacent agents, and task-agnostic hardware scheduling.<n>We present a framework that addresses all three simultaneously through: (1) Differential Neural Caching, which stores intermediate layer activations and computes only the input deltas; (2) Criticality-Based Action Space Pruning, which organizes agents into priority tiers and reduces coordination complexity from O(n2) to O(n log n) with less than 6% optimal
  arXiv  Detail & Related papers  (2026-04-22T02:54:35Z)
• SABER: A Stealthy Agentic Black-Box Attack Framework for Vision-Language-Action Models [55.48982086345088]
  SABER is an agent-centric approach for automatically generating instruction-based adversarial attacks on VLA models.<n>We show that SABER reduces task success by 20.6%, increases action-sequence length by 55%, and raises constraint violations by 33%.
  arXiv  Detail & Related papers  (2026-03-26T01:56:01Z)
• Towards a Science of Scaling Agent Systems [79.64446272302287]
  We formalize a definition for agent evaluation and characterize scaling laws as the interplay between agent quantity, coordination structure, modelic, and task properties.<n>We derive a predictive model using coordination metrics, that cross-validated R2=0, enabling prediction on unseen task domains.<n>We identify three effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead, and (2) a capability saturation: coordination yields diminishing or negative returns once single-agent baselines exceed 45%.
  arXiv  Detail & Related papers  (2025-12-09T06:52:21Z)
• DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning [61.10299147201369]
  This paper introduces a novel autonomous RL approach, called DigiRL, for training in-the-wild device control agents. We build a scalable and parallelizable Android learning environment equipped with a VLM-based evaluator. We demonstrate the effectiveness of DigiRL using the Android-in-the-Wild dataset, where our 1.3B VLM trained with RL achieves a 49.5% absolute improvement.
  arXiv  Detail & Related papers  (2024-06-14T17:49:55Z)
• Real-Time Model-Free Deep Reinforcement Learning for Force Control of a Series Elastic Actuator [56.11574814802912]
  State-of-the art robotic applications utilize series elastic actuators (SEAs) with closed-loop force control to achieve complex tasks such as walking, lifting, and manipulation. Model-free PID control methods are more prone to instability due to nonlinearities in the SEA. Deep reinforcement learning has proved to be an effective model-free method for continuous control tasks.
  arXiv  Detail & Related papers  (2023-04-11T00:51:47Z)
• Improving Efficiency in Large-Scale Decentralized Distributed Training [58.80224380923698]
  We propose techniques to accelerate (A)D-PSGD based training by improving the spectral gap while minimizing the communication cost. We demonstrate the effectiveness of our proposed techniques by running experiments on the 2000-hour Switchboard speech recognition task and the ImageNet computer vision task.
  arXiv  Detail & Related papers  (2020-02-04T04:29:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.