HarmonE: A Self-Adaptive Approach to Architecting Sustainable MLOps

2025-05-19

AI Summary

Machine learning systems (MLS) in dynamic environments suffer from multi-dimensional sustainability degradation due to data drift and model deterioration. Method: This paper proposes an adaptive MLOps architecture integrating technical, economic, environmental, and social sustainability objectives—first explicitly embedding sustainability goals into the MLOps design phase. It establishes a synergistic optimization framework combining the MAPE-K (Monitor-Analyze-Plan-Execute-Knowledge) control loop with digital twin technology, and introduces a novel joint monitoring mechanism for energy consumption, prediction accuracy, and data drift, coupled with a threshold-driven adaptive policy triggering engine. Results: Evaluated on an intelligent transportation digital twin platform, the architecture reduces redundant retraining energy consumption significantly, sustains prediction accuracy above 92%, and simultaneously satisfies carbon footprint constraints and real-time responsiveness requirements—providing a scalable methodological foundation for sustainable MLS.

Technology Category

• Machine Learning: Scalability of ML Systems
• Humans and AI: Human-in-the-loop Machine Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

Machine Learning Enabled Systems (MLS) are becoming integral to real-world applications, but ensuring their sustainable performance over time remains a significant challenge. These systems operate in dynamic environments and face runtime uncertainties like data drift and model degradation, which affect the sustainability of MLS across multiple dimensions: technical, economical, environmental, and social. While Machine Learning Operations (MLOps) addresses the technical dimension by streamlining the ML model lifecycle, it overlooks other dimensions. Furthermore, some traditional practices, such as frequent retraining, incur substantial energy and computational overhead, thus amplifying sustainability concerns. To address them, we introduce HarmonE, an architectural approach that enables self-adaptive capabilities in MLOps pipelines using the MAPE-K loop. HarmonE allows system architects to define explicit sustainability goals and adaptation thresholds at design time, and performs runtime monitoring of key metrics, such as prediction accuracy, energy consumption, and data distribution shifts, to trigger appropriate adaptation strategies. We validate our approach using a Digital Twin (DT) of an Intelligent Transportation System (ITS), focusing on traffic flow prediction as our primary use case. The DT employs time series ML models to simulate real-time traffic and assess various flow scenarios. Our results show that HarmonE adapts effectively to evolving conditions while maintaining accuracy and meeting sustainability goals.

Problem

• Ensuring sustainable performance of Machine Learning Enabled Systems (MLS) in dynamic environments
• Addressing technical, economical, environmental, and social sustainability in MLOps
• Reducing energy and computational overhead from frequent retraining in MLS

Innovation

• Self-adaptive MLOps using MAPE-K loop
• Runtime monitoring of sustainability metrics
• Digital Twin validates traffic flow prediction