Dynamic Executable Selection for Power Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face impracticality in adapting to varying operating environments due to the need for manual configurations and the inability to consider all possible variations in actual execution environments, leading to inefficient performance and power consumption.
Innovation Solution
Dynamic program evaluation method that tests multiple executable variations to select the most suitable one based on runtime sensor data, automatically adapting to changes in hardware and software environments to optimize performance and power efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple versions of source codes are provided for the same task, then the system can adapt to different operating environments, but the complexity of system configuration increases and manual intervention is required
Solution Approach 1:
The system automatically selects and executes the most suitable executable version based on runtime sensor data and environmental conditions, eliminating the need for manual configuration. The multi-executable system self-adapts by monitoring its own operating environment and autonomously determining which executable to use, thus providing self-service functionality.
Solution Approach 2:
The system changes operational parameters dynamically by selecting different executable versions based on detected environmental parameters (temperature, power source, performance metrics). Instead of fixed configuration, the system adapts by changing which executable is active based on real-time parameter monitoring.
2Use of energy by stationary object
If manual configuration is used to specify which executable to use, then the system can control power consumption and performance, but the ease of operation deteriorates and the process becomes tedious
Solution Approach 1:
The system automatically monitors power consumption during test runs of different executables and selects the most power-efficient version without user intervention. Power management becomes a self-service function where the system autonomously optimizes its energy usage based on runtime measurements.
Solution Approach 2:
The system implements feedback loops by measuring power consumption and performance metrics during executable test runs, then using this feedback to automatically select the optimal executable. The feedback mechanism enables autonomous decision-making about resource usage without manual configuration.
3Measurement precision
If test runs are conducted for multiple executables to evaluate performance, then the accuracy of executable selection improves, but the loss of time increases due to multiple executions
Solution Approach 1:
The system performs a limited number of test runs with a subset of executables rather than exhaustively testing all possible versions. By selecting and testing only the most relevant candidates based on initial conditions, the system achieves sufficient measurement precision without excessive time investment.
Solution Approach 2:
The system performs preliminary filtering or selection of candidate executables before conducting full test runs. By pre-identifying the most promising candidates based on available information, the system reduces the scope of subsequent testing while maintaining selection accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and apparatus to maintain a plurality of executables for a task in a device are described. Each executable may be capable of performing the task in response to a change in an operating environment of the device. Each executable may be executed to perform a test run of the task. Each execution can consume an amount of power under the changed operating environment in the device. One of the executables may be selected to perform the task in the future based on the amounts of power consumed for the test runs of the task. The selected one executable may require no more power than each of remaining ones of the executables.