Resource-Utility Function Mapping for Real-Time Task Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optimizing resource allocation among multiple tasks in real-time systems is challenging due to the complexity of maximizing utility with limited resources, requiring high computational effort and inefficient processes.
Innovation Solution
A method and apparatus that determine parameters of a resource-utility function based on task states to efficiently allocate resources by mapping task states to parameters using a trained model, allowing for fast and efficient resource distribution without exceeding predefined thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional optimization methods are used to maximize utility with limited resources, then resource allocation accuracy is improved, but computational effort and processing time increase significantly
Solution Approach 1:
The patent pre-calculates and stores utility values for different resource allocation scenarios in a lookup table before runtime. During actual resource allocation, the system simply queries this pre-computed table rather than performing optimization calculations, thereby achieving accurate resource allocation without real-time computational overhead.
Solution Approach 2:
The patent creates a simplified model or lookup table that copies essential utility information from complex optimization scenarios. This copied representation allows the system to retrieve allocation decisions quickly without re-running complex optimization algorithms, maintaining accuracy while reducing computational effort.
2Productivity
If complex optimization algorithms are applied to distribute resources among multiple tasks, then utility maximization is improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent divides the resource allocation problem into discrete scenarios and pre-computes utility values for each scenario independently. By segmenting the complex optimization problem into manageable lookup entries, the system avoids the need for complex real-time optimization algorithms while still achieving utility maximization.
Solution Approach 2:
The patent transforms the continuous optimization problem into a discrete lookup operation by parameterizing different resource allocation scenarios. This parameterization allows the system to switch between pre-computed solutions based on current conditions without performing complex calculations, thereby reducing computational complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method to distribute resources among a group of tasks is provided. The method includes determining at least one parameter of a resource-utility function for a task of the group of tasks based on a task state of the task and determining a resource to be consumed by the task based on the resource-utility function.