Neural Network Hardware Selection by Brown Energy and Green Power Ratio
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for selecting hardware to execute neural network models do not consider the consumption amount of brown energy, limiting the optimization of these models based on renewable energy usage.
Innovation Solution
An execution hardware determination method that includes query reception, search processing for suitable neural network models, preliminary calculation of energy consumption using an energy prediction model, and determination of hardware based on green power ratio to select hardware that minimizes brown energy consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If hardware selection is based on traditional performance metrics only, then execution speed and model accuracy are improved, but brown energy consumption increases without optimization
Solution Approach 1:
The patent changes the selection parameters for hardware from traditional performance metrics alone to include green power ratio as a new dimension. The determination unit selects execution hardware based on both performance requirements and the proportion of renewable energy, transforming the hardware selection criteria to optimize for sustainability while maintaining execution efficiency.
Solution Approach 2:
The system incorporates feedback mechanisms by continuously monitoring and comparing green power ratios of different hardware options. The determination unit uses this feedback information to dynamically select the most environmentally friendly hardware that still meets performance requirements, creating a closed-loop decision-making process.
2Loss of energy
If hardware selection considers green power ratio, then brown energy consumption is reduced, but hardware selection complexity increases
Solution Approach 1:
The system performs preliminary action by pre-calculating and storing green power ratios for various hardware devices in advance. This pre-computed information is then readily available to the determination unit, eliminating the need for complex real-time calculations and simplifying the hardware selection process while still achieving energy optimization.
3Ease of manufacture
If existing hardware selection methods are used, then implementation is simple, but sustainability optimization is not achieved
Solution Approach 1:
The determination unit is designed with multi-functionality, serving both traditional performance-based hardware selection and the new green power ratio-based selection. This universal approach allows the system to maintain simplicity by using a unified selection mechanism that can handle multiple criteria (performance + sustainability) without requiring separate complex systems.
Data Source
AI summary
Hardware suitable for execution of a neural network model can be selected from the viewpoint of a consumption amount of brown energy. A computer determines execution hardware, which is hardware that executes a neural network model. The execution hardware determination method includes: query reception processing for reading a user query in which a use case of the neural network model; search processing for searching a model database for a standard model, which is a neural network model that satisfies most of the constraint of the candidate hardware and the performance condition; preliminary calculation processing for inputting to an energy prediction model a performance metric of the standard model and the constraint of the candidate hardware based on the user query; and determination processing for determining the execution hardware that executes a work load which is a proportion of renewable energy to energy supplied to the candidate hardware.