Switching Frequency as FPGA Monitor: Studying Degradation and Ageing Prognosis at Large Scale

TL;DR

This study analyzes FPGA switching frequency degradation in a large-scale deployment, demonstrating continuous aging trends and using machine learning to accurately forecast future degradation over two months.

Contribution

It provides the first large-scale statistical analysis of FPGA degradation in operational environments and introduces ML-based forecasting for predictive maintenance.

Findings

Degradation of switching frequency is continuous and measurable.

Machine learning models can predict future degradation with high accuracy.

Identified specific FPGA areas highly impacted by aging.

Abstract

The growing deployment of unhardened embedded devices in critical systems demands the monitoring of hardware ageing as part of predictive maintenance. In this paper, we study degradation on a large deployment of 298 naturally aged FPGAs operating in the European XFEL particle accelerator. We base our statistical analyses on 280 days of in-field measurements and find a generalized and continuous degradation of the switching frequency across all devices with a median value of 0.064%. The large scale of this study allows us to localize areas of the deployed FPGAs that are highly impacted by degradation. Moreover, by training machine learning models on the collected data, we are able to forecast future trends of frequency degradation with horizons of 60 days and relative errors as little as 0.002% over an evaluation period of 100 days.