Accurate and Stable Run-Time Power Modeling for Mobile and Embedded CPUs

Abstract

Abstract-Modern mobile and embedded devices are required to be increasingly energy-efficient while running more sophisticated tasks, causing the CPU design to become more complex and employ more energy-saving techniques. This has created a greater need for fast and accurate power estimation frameworks for both run-time CPU energy management and design-space exploration. We present a statistically rigorous and novel methodology for building accurate run-time power models using Performance Monitoring Counters (PMCs) for mobile and embedded devices, and demonstrate how our models make more efficient use of limited training data and better adapt to unseen scenarios by uniquely considering stability. Our robust model formulation reduces multicollinearity, allows separation of static and dynamic power, and allows a 100× reduction in experiment time while sacrificing only 0.6% accuracy. We present a statistically detailed evaluation of our model, highlighting and addressing the problem of heteroscedasticity in power modeling. We present software implementing our methodology and build power models for ARM Cortex-A7 and Cortex-A15 CPUs, with 3.8% and 2.8% average error, respectively. We model the behavior of the nonideal CPU voltage regulator under dynamic CPU activity to improve modeling accuracy by up to 5.5% in situations where the voltage cannot be measured. To address the lack of research utilizing PMC data from real mobile devices, we also present our data acquisition method and experimental platform software. We support this work with online resources including software tools, documentation, raw data and further results.

A Novel Easy-to-construct Power Model for Embedded and Mobile Systems - Using Recursive Neural Nets to Estimate Power Consumption of ARM-based Embedded Systems and Mobile Devices

“…The authors later continued to enhance their model with better power estimation for the GPU (Kim and Chung, 2013), and the display (Kim et al, 2015). More recently, several new models were also published, either for the system as a whole, as it was the case for (Di Nucci et al, 2017a), (Shukla et al, 2016) and (Chowdhury and Hindle, 2016), or just the Central Processing Unit (CPU) like the models presented in (Yoon et al, 2017) and (Walker et al, 2017). Some of them even offered better accuracy like (Yoon et al, 2017) and proved that software profiling is practically as accurate and reliable as hardware profiling (Di Nucci et al, 2017b).…”

Thermally-aware composite run-time CPU power models

“…Previously, top-down regression-based models using performance monitoring counters (PMCs) as inputs have been widely shown to be effective in estimating CPU power [2]- [13]. Topdown approaches use power measurements from real devices and predict this measured power using metrics.…”

“…Obtaining trustworthy simulation results is a key challenge in research and using tools without understanding their limitations can lead to incorrect research conclusions. While the flexibility of bottom-up tools is sometimes required, there are many cases where a topdown model (built and validated on a real hardware platform) can be used to provide an accurate and trusted reference, as highlighted in [13].…”

“…A recent work [13] overcomes this and significantly improves on the state-ofthe-art by presenting a statistically-rigorous methodology that ensures stability, deals with heteroscedasticity and achieves a low error on two mobile CPUs of differing microarchitecture. However, the effects of temperature on power consumption are not considered and the model could be improved by further decomposing the estimated power consumption, and therefore providing more information on how the power is being consumed.…”

“…However, in order to make these trade-offs effectively, run-time knowledge of how power is being consumed is essential. A fast and accurate power model, that is able to predict the current power consumption at runtime, can therefore be used with a run-time manager to make significant energy savings [1].Previously, top-down regression-based models using performance monitoring counters (PMCs) as inputs have been widely shown to be effective in estimating CPU power [2]- [13]. Topdown approaches use power measurements from real devices and predict this measured power using metrics.…”