Computing Resource Allocation for Edge Task QoS and Power Saving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication networks do not provide computing power resources for user equipment, limiting the ability to efficiently allocate resources for computing tasks and ensuring quality of service requirements.
Innovation Solution
A method and apparatus for allocating computing power and communication resources by network elements, including computing power control, session management, and policy control network elements, based on task information to meet computing power and communication QoS policies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If terminal devices perform computing tasks locally, then computing autonomy is maintained, but device complexity and power consumption increase
Solution Approach 1:
The patent segments the computing task execution between terminal devices and edge computing servers. The terminal device segments heavy computing tasks and offloads them to edge servers, while retaining local execution for simpler tasks. This segmentation allows the terminal to maintain computing autonomy for critical functions while reducing power consumption by leveraging external computing resources.
Solution Approach 2:
The patent introduces an edge computing server as an intermediary between the terminal device and the cloud. This intermediary provides computing power resources to the terminal device, enabling the terminal to offload computing tasks without directly connecting to distant cloud infrastructure. The intermediary reduces the terminal's power consumption while maintaining computing autonomy through local edge collaboration.
2Reliability
If computing power resources are allocated to terminal devices, then service quality improves, but network complexity increases
Solution Approach 1:
The patent creates a universal computing power resource allocation framework where edge computing servers provide standardized computing resources to multiple terminal devices. The allocation mechanism handles diverse computing tasks (AI inference, video processing, gaming) through a unified resource management system. This multi-functionality improves service quality across different applications while avoiding the need for separate specialized systems for each task type.
Solution Approach 2:
The patent dynamically adjusts allocation parameters such as computing resource quantity, task priority levels, and quality of service thresholds based on real-time network conditions and terminal device requirements. By changing these parameters adaptively, the system maintains high service quality while optimizing network resource utilization and managing complexity through standardized parameter control mechanisms.
3Manufacturing precision
If more computing power resources are allocated, then task completion accuracy improves, but resource allocation cost increases
Solution Approach 1:
The patent implements partial action by allocating computing power resources dynamically based on task requirements rather than providing full resources continuously. The system allocates just enough computing power to achieve the required task completion accuracy for each specific scenario. For example, simple tasks receive minimal resources while complex AI inference tasks receive adequate but not excessive resources, optimizing the balance between accuracy and cost.
Solution Approach 2:
The patent changes resource allocation parameters adaptively based on task characteristics and quality requirements. The system adjusts computing resource allocation levels, task scheduling priorities, and quality of service parameters to match actual needs. This dynamic parameter adjustment ensures high task completion accuracy when necessary while reducing resource allocation costs during normal operations, achieving optimal cost-performance balance.
Data Source
AI summary
A method for allocating a resource for a computing task includes obtaining, by a computing power control network element, a first computing task request, where the first computing task request is used to request to allocate a computing power resource for a target computing task of a terminal device, and the first computing task request includes task information of the target computing task; and allocating, by the computing power control network element, the computing power resource for the target computing task based on the task information.