Edge Server Deployment via Virtual Geometric Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for deploying edge servers are limited as they only consider a single objective and are deployed at predetermined sites, missing opportunities to improve network delay, cost, and user performance by not accounting for all practical factors.
Innovation Solution
A method and apparatus that map IP addresses of target users into a virtual geometric space, cluster users based on coordinates, and determine edge server locations to optimize deployment across geographic and network locations, considering multiple factors for better service quality and cost-effectiveness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If edge servers are deployed only at given candidate server sites, then the deployment process is simple, but the network delay optimization opportunity is lost
Solution Approach 1:
The patent transforms the traditional one-dimensional candidate site selection into a multi-dimensional optimization problem by considering both geographic location and network topology dimensions. Users are mapped to virtual geometric space coordinates, and edge server locations are determined by clustering users in this multi-dimensional space, enabling simultaneous optimization of network delay and deployment effectiveness.
Solution Approach 2:
The system automatically determines optimal edge server locations by clustering users based on their coordinates in virtual geometric space. The algorithm self-organizes user data and autonomously identifies optimal deployment locations without requiring manual selection from predetermined candidates, thereby discovering unknown better sites.
2Device complexity
If edge servers are deployed optimizing for only one single objective, then the optimization process is simple, but the overall service quality is compromised
Solution Approach 1:
The patent changes the optimization parameters from single-objective to multi-objective by introducing weight coefficients for different factors. The location determination formula incorporates multiple parameters including user distribution, network delay, and deployment cost, allowing flexible adjustment of optimization priorities through parameter changes while maintaining comprehensive service quality.
Solution Approach 2:
The clustering algorithm serves multiple functions simultaneously: it groups users by geographic proximity, identifies optimal edge server locations, and provides tradeoff curves for different deployment scenarios. This multi-functional approach eliminates the need for separate optimization processes for each objective.
3Reliability
If more edge servers are deployed to improve user performance, then service quality improves, but the deployment cost increases
Solution Approach 1:
Instead of deploying edge servers to serve all users uniformly, the patent applies partial action by clustering users and deploying servers only to critical regions where user density and performance requirements demand it. The algorithm identifies the minimum necessary number of servers to achieve acceptable service quality levels, avoiding excessive deployment costs.
Solution Approach 2:
The patent introduces cost parameters and service quality thresholds as adjustable parameters in the optimization formula. By changing these parameters, the system can balance between deploying more servers for higher performance and deploying fewer servers for cost efficiency, providing tradeoff curves for different budget scenarios.
Data Source
AI summary
A method and apparatus for deploying edge servers are provided. The method includes: obtaining IP addresses of target users for the edge server to be deployed; mapping the IP addresses of the target users into a virtual geometric space; clustering the target users based on coordinates of the target users in the virtual geometric space, to obtain a plurality of user groups; and determining locations of the edge servers corresponding to the plurality of user groups based on the coordinates of the target users in the plurality of user groups.