CARINA: Carbon-Aware Execution of Recurrent Industrial Analytics

2026-05-23

AI Summary

This work addresses the widespread lack of energy and carbon awareness in industrial-scale repetitive analytics and machine learning workflows, which often results in excessive energy consumption and high carbon footprints. The authors propose CARINA, a novel framework that introduces carbon-aware scheduling into such pipelines without requiring device-level carbon metering. CARINA estimates emissions using grid emission factors and integrates lightweight runtime monitoring, peak electricity demand detection, and adaptive execution strategies to avoid the unintended increase in total energy use that can arise from naive throttling. Transparent reporting is enabled through a local visualization dashboard. Evaluated on two automotive OEM database generation tasks, CARINA reduces full-cycle energy consumption by approximately 9% with only a 7% runtime overhead, achieving carbon emission reductions of 21.8 kg CO₂e and 33.2 kg CO₂e, respectively.

Abstract

Recurring industrial analytics and machine-learning workflows are becoming a major computational burden in modern engineering practice. Large parametric database generation, scheduled model retraining, repeated evaluation pipelines, and extensive hyperparameter exploration can demand hundreds of runtime hours and tens of kilowatt-hours per refresh cycle, yet these workloads are rarely executed with explicit energy-awareness. We present CARINA (Carbon-Aware Recurrent Industrial Analytics), a measurement-and estimation framework for energy-aware and carbon-aware execution of recurrent analytics. The framework combines lightweight run-level and step-level instrumentation, peak time-aware execution control, and local dashboard reporting. The method estimates energy load as the primary objective and translates it to carbon emissions using a local grid emission factor, enabling use even when direct device level carbon metrology is unavailable. We evaluate the framework using two automotive OEM database-generation workflows. The first required 1.48 million scenarios, 180.30 h, and 48.67 kWh; the second required 3.66 million scenarios, 274.75 h, and 74.16 kWh (corresponding to approximately 21.8 kg CO2e and 33.2 kg CO2e, respectively). Preliminary policy analysis suggests that peak-aware off-hours boosting can reduce full-cycle energy load by about 9% with roughly 7% runtime overhead, while naive throttling can increase total energy through overhead effects.

Problem

• carbon-aware computing
• industrial analytics
• energy consumption
• machine learning workflows
• carbon emissions

Innovation

• carbon-aware computing
• energy estimation
• industrial analytics
• peak-aware scheduling
• emission factor