Network Slice Resource Allocation via Utility Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication network architectures lack flexibility and are insufficient to support the diverse requirements of different service scenarios, such as mobility, security, and quality of service, leading to inefficient resource allocation and computational complexity in managing multiple network slices.
Innovation Solution
A method is introduced where a network device determines the utility values of network slices based on resource demand and priority, reallocating system resources only when the difference in utility values between slices exceeds a threshold, thereby reducing the frequency of resource updates and ensuring fairness in resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If frequent resource updates are performed to ensure fair resource allocation among network slices, then resource allocation fairness is improved, but computational complexity increases
Solution Approach 1:
The patent implements periodic resource updates triggered by threshold conditions rather than continuous updates. The network device monitors resource allocation and only performs reallocation when the difference in utility values exceeds a predetermined threshold, creating a periodic action pattern that reduces computational complexity while maintaining fairness.
Solution Approach 2:
The patent employs feedback mechanisms by calculating utility values for each network slice based on resource allocation and demand, comparing these values against thresholds, and using the comparison results to trigger reallocation decisions. This feedback loop ensures fairness is maintained only when necessary, reducing unnecessary computational overhead.
2Productivity
If continuous resource reallocation is performed to optimize system resource utilization, then resource utilization efficiency is improved, but the frequency of resource adjustments increases
Solution Approach 1:
The system performs resource reallocation periodically based on threshold-triggered events rather than continuously. Updates occur only when utility value differences exceed predetermined thresholds, reducing the frequency of adjustments while maintaining optimal resource utilization efficiency through timely reallocations.
3Device complexity
If utility value threshold triggering is used to reduce resource update frequency, then computational complexity is reduced, but resource allocation responsiveness may decrease
Solution Approach 1:
The patent changes the parameter for triggering resource updates from continuous monitoring to threshold-based utility value comparison. By monitoring whether utility value differences exceed predetermined thresholds rather than continuously optimizing, the system reduces computational complexity while maintaining responsive allocation when actual fairness issues arise.
Data Source
Figure 1~2
Figure 3~4
AI summary
Embodiments of the present disclosure relate to a method, a device, and a computer-readable storage medium for allocating system resources among a plurality of network slices in a communication network. In the communication network, a network device determines utility values of two network slices. A network slice corresponds to a set of network functions of a set of services served thereby. A utility value of the network slice indicates the level of demand for system resources by the set of services associated with the network slice. The network device determines a difference between the utility values of the two network slices. If the difference exceeds a threshold utility value, the network device re-allocates system resources among the plurality of network slices in the communication network. According to the embodiments, the communication network updates the system resources allocated for the plurality of network slices only when the difference between the utility values of the two network slices exceeds a certain threshold. Hence, the complexity of updating system resource allocation among the plurality of network slices may be reduced by the method according to the embodiments of the present disclosure.