Data Dependent Energy Modeling for Worst Case Energy Consumption Analysis

摘要: Safely meeting Worst Case Energy Consumption (WCEC) criteria requires accurate energy modeling of software. We investigate the impact instruction operand values upon consumption in cacheless embedded processors. Existing instruction-level models typically use measurements from random input data, providing estimates unsuitable for safe WCEC analysis.We examine probabilistic distributions instructions and propose a model composing sequences using distributions, enabling analysis on program basic blocks. The worst case is predicted with statistical analysis. Further, we verify that benchmarks can be characterised as distribution, compare our proposed technique other methods estimating consumption.