ReadyPower: A Reliable, Interpretable, and Handy Architectural Power Model Based on Analytical Framework

2025-12-16

AI Summary

Existing architectural-level power modeling approaches suffer from low accuracy of analytical models (e.g., McPAT) and poor reliability, limited interpretability, and deployment challenges of machine learning–based methods. To address these issues, this paper proposes a novel hybrid power modeling methodology that integrates analytical modeling with three-tier parameter calibration—architectural, implementation, and process levels—built upon an extended McPAT framework. Unlike black-box ML models, our approach preserves physical interpretability, usability, and industrial deployability, and supports open-source RISC-V cores including BOOM and XiangShan. Experimental evaluation demonstrates that our model achieves over 20% reduction in mean absolute percentage error (MAPE) and improves the correlation coefficient *R* by more than 0.2 across two representative processor families. It significantly outperforms state-of-the-art ML-based baselines while ensuring high accuracy, trustworthiness, and engineering practicality.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: (Computational) Cognitive Architectures
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Power is a primary objective in modern processor design, requiring accurate yet efficient power modeling techniques. Architecture-level power models are necessary for early power optimization and design space exploration. However, classical analytical architecture-level power models (e.g., McPAT) suffer from significant inaccuracies. Emerging machine learning (ML)-based power models, despite their superior accuracy in research papers, are not widely adopted in the industry. In this work, we point out three inherent limitations of ML-based power models: unreliability, limited interpretability, and difficulty in usage. This work proposes a new analytical power modeling framework named ReadyPower, which is ready-for-use by being reliable, interpretable, and handy. We observe that the root cause of the low accuracy of classical analytical power models is the discrepancies between the real processor implementation and the processor's analytical model. To bridge the discrepancies, we introduce architecture-level, implementation-level, and technology-level parameters into the widely adopted McPAT analytical model to build ReadyPower. The parameters at three different levels are decided in different ways. In our experiment, averaged across different training scenarios, ReadyPower achieves >20% lower mean absolute percentage error (MAPE) and >0.2 higher correlation coefficient R compared with the ML-based baselines, on both BOOM and XiangShan CPU architectures.baselines, on both BOOM and XiangShan CPU architectures.

Problem

• Develops a reliable analytical power model for processors
• Improves accuracy over classical and ML-based power models
• Enables early power optimization and design space exploration

Innovation

• Enhances McPAT model with multi-level parameters for accuracy
• Introduces architecture, implementation, and technology parameters to bridge discrepancies
• Achieves lower error and higher correlation than ML-based baselines