Computing Native Network Dynamic Resource Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless network AI schemes fail to concurrently support network and computing services, leading to inefficient resource utilization and the need for additional hardware investments, while also lacking dynamic scheduling and energy-efficient computing configurations.
Innovation Solution
A computing native network mechanism that dynamically allocates and schedules computing resources within existing wireless network devices, allowing for concurrent support of network and computing services without additional hardware, and incorporates cloud computing for enhanced resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If separate computing cards are added to base stations to support computing services, then computing service capability is improved, but device complexity and hardware investment increase
Solution Approach 1:
The base station is designed to perform both network service and computing service using the same hardware resources. The computing service is implemented by utilizing existing base station computing resources rather than adding dedicated computing cards, making the base station a multi-functional device that can handle both communication and computing tasks concurrently.
Solution Approach 2:
The patent merges network service functionality and computing service functionality into a single integrated system. By combining these services and sharing underlying hardware resources (CPU, memory, storage), the system avoids the need for separate computing cards while achieving concurrent support for both service types.
2Reliability
If base station computing resources are fully allocated to network services, then network service quality is improved, but computing resource utilization deteriorates when load is low
Solution Approach 1:
The system implements dynamic resource allocation where the base station's computing resources are flexibly adjusted between network services and computing services based on real-time load conditions. When network load is low, more computing resources are allocated to computing services; when network load is high, resources are dynamically shifted back to network services, maintaining both service qualities.
Solution Approach 2:
The patent changes the allocation parameter of computing resources dynamically based on network load conditions. By monitoring network traffic and adjusting the proportion of computing resources allocated to different services, the system optimizes both network service quality and computing resource utilization according to actual operational needs.
3Productivity
If computing resources are dynamically shared between network services and computing services, then resource utilization is improved, but scheduling complexity increases
Solution Approach 1:
The computing resources are segmented into different allocable units that can be dynamically assigned to network services or computing services. The scheduling mechanism divides resources into manageable segments that can be independently allocated, making the dynamic sharing process more controllable and less complex than managing monolithic resources.
Data Source
AI summary
A computing provider side device and a computing scheduling management side device in a wireless communication system are proposed. The computing scheduling management side device can collect computing related information about shareable computing from the computing provider side, and information about an application on a computing consumer side that needs to utilize the computing, and perform computing scheduling for the application based on the information. The computing provider side device can determine and provide its shareable computing for being scheduled for the applications on the provider side, and such shareable computing also can be released so as to be reused by the computing provider side device.