Virtualized Base Station Route Switching for Shared Accelerator Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently managing increased processing loads in virtualized base stations due to 5G advancements, particularly in distributing CPU processes and reducing power consumption while maintaining low latency, without significantly increasing costs.
Innovation Solution
A wireless communication system where virtualized base stations are controlled by a RAN controller, utilizing accelerators (ACC) in select computers, and dynamically switching routes based on ACC and traffic information to optimize processing capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the number of virtualized base stations is increased to handle large volume traffic, then the processing capability is improved, but the power consumption and device complexity increase
Solution Approach 1:
The patent implements dynamic route switching between direct and relay routes based on real-time accelerator availability and traffic conditions. This allows the system to adaptively adjust the number of active virtualized base stations, maintaining high processing capability while minimizing power consumption by utilizing relay routes through accelerators only when necessary
Solution Approach 2:
The accelerator functions as a multi-purpose resource that can serve multiple virtualized base stations through relay communication. Instead of dedicating separate accelerators to each virtualized base station, a single accelerator can assist multiple CUs/DUs by relaying traffic, thereby reducing overall power consumption while maintaining processing capability
2Productivity
If accelerators are installed in all CUs and DUs to reduce CPU processing load, then the processing efficiency is improved, but the cost increases significantly
Solution Approach 1:
The patent merges the accelerator resources into a shared pool that serves multiple CUs and DUs through relay routes. Instead of each CU/DU having its own dedicated accelerator, the system combines accelerator resources centrally and utilizes them via relay communication, significantly reducing the total number of accelerators needed while maintaining processing efficiency
Solution Approach 2:
The relay route introduces an intermediary accelerator that mediates between the CPU and the network traffic. Instead of requiring accelerators in every CU/DU, a single accelerator can act as an intermediary to offload processing tasks from multiple CUs/DUs, reducing cost while improving processing efficiency
3Productivity
If route switching is implemented dynamically based on accelerator information, then the load balancing is improved, but the system complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where the RIC continuously collects accelerator information and traffic data from CUs and DUs, processes this information to determine optimal routes, and sends route switching instructions back to the CUs and DUs. This automated feedback loop enables effective load balancing without requiring complex manual configuration or coordination
Data Source
AI summary
Each of virtualized base stations is configured by connecting respective units of CU, DU, and RUs. An ACC is installed at least a part of computers that virtualize the respective units. The respective units notify the RAN controller of ACC information of computers and traffic information. The RAN controller comprises a route calculation unit 101 that calculates an optimal route based on the ACC information and traffic information of the respective units, and a route notification unit 102 that notifies the respective units of information on the optimal route, and the respective units switch the route between the respective units based on the notified route information.