Network Slicing via Metadata Classification and Resource Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network technologies face challenges in efficiently managing and allocating resources to meet the diverse performance requirements of multiple services coexisting on a single physical network, leading to suboptimal bandwidth utilization and increased maintenance costs.
Innovation Solution
A network slicing method that uses metadata information from traffic flows to determine classification and slicing policies, allocating dedicated virtual resources to each service, ensuring efficient network utilization and dynamic adjustment to meet changing service needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple services are carried on a single physical network to reduce management costs, then network utilization improves, but service performance guarantee deteriorates
Solution Approach 1:
The patent divides a single physical network into multiple virtual network slices, each dedicated to specific service types with different performance requirements. This segmentation allows simultaneous optimization for different services (e.g., high bandwidth for video, low delay for voice) while maintaining efficient resource utilization through shared physical infrastructure.
Solution Approach 2:
The patent applies different quality parameters and resource allocation strategies to different network slices based on their specific service requirements. Each slice receives customized performance guarantees (bandwidth, delay, jitter) tailored to its service type, while the overall network maintains high utilization through coordinated resource management across all slices.
2Reliability
If differentiated service and high bandwidth are used to guarantee quality of service, then service performance improves, but bandwidth utilization deteriorates
Solution Approach 1:
The patent implements dynamic resource allocation within each network slice, allowing bandwidth and other resources to be flexibly adjusted based on real-time traffic demands and service priorities. This dynamic management ensures quality of service guarantees are met while minimizing wasted bandwidth through adaptive resource distribution.
Solution Approach 2:
The patent incorporates feedback mechanisms that monitor network slice performance and resource usage, enabling the system to adjust resource allocation in response to changing conditions. This feedback loop ensures service quality is maintained while optimizing bandwidth utilization by reallocating resources from underutilized to high-demand slices.
3Productivity
If centralized flow control technology is used to achieve optimal network performance, then network utilization improves, but ability to meet diverse service requirements deteriorates
Solution Approach 1:
The patent segments the network into multiple virtual slices, each with dedicated flow control policies tailored to specific service types. This allows the system to achieve optimal performance for each service category (e.g., TCP-friendly flow control for web traffic, rate-limiting for signaling) while maintaining high overall network utilization through coordinated management.
Solution Approach 2:
The patent applies different flow control strategies and performance parameters to different network slices based on their specific service requirements. Each slice receives customized flow management (e.g., priority queuing for voice, best-effort for data) that optimizes performance for its service type while the overall system maintains high utilization.
Data Source
AI summary
This application discloses a network slicing method. The method includes obtaining metadata information of each of multiple traffic flows, determining a classification policy according to metadata information of the multiple traffic flows, determining a network slicing policy according to the classification policy, and the network slicing policy includes a mapping relationship between a traffic flow type and a network slice and a slice weight of the network slice, and determining a resource node policy according to the network slicing policy, where the resource node policy includes a mapping relationship between at least one resource node in the network and the at least one network slice and a resource weight of each resource node, and the resource weight of the resource node indicates a ratio of a resource of each of network slices corresponding to the resource node to a resource of the resource node.