Edge Computing Application Relocation Prediction for V2X
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing-based application relocation methods in multi-access edge computing (MEC) systems face delays in detecting UE bearer changes, creating new application instances, and transferring context, leading to packet loss and service interruptions, particularly in vehicle-to-everything (V2X) services with frequent high-speed movements.
Innovation Solution
Predicting a target edge computing host in advance and preparing the necessary resources, such as creating an application instance and transmitting context information, to minimize delays during the actual relocation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If application relocation is triggered by UE bearer change detection, then the relocation process can be initiated, but the detection delay and subsequent processing delays cause packet loss and service interruption
Solution Approach 1:
The patent applies preliminary action by predicting the target MEC platform in advance based on UE movement trajectory and network topology, then pre-establishing the application instance and context transfer path before actual relocation is needed. This eliminates the detection delay and processing delays that occur with reactive approaches, ensuring continuous service during UE mobility.
2Productivity
If application instance creation and context transfer are performed sequentially, then the relocation process is simplified, but the sequential processing causes delays that lead to packet loss
Solution Approach 1:
The patent performs preliminary actions by pre-creating the application instance on the target MEC platform and pre-establishing the context transfer path before actual relocation. This parallel preparation eliminates sequential processing delays and ensures that when relocation is triggered, the target environment is already ready, significantly reducing processing time and preventing packet loss.
Solution Approach 2:
The patent ensures continuity of useful action by maintaining the application instance and context transfer path in a ready state throughout the prediction period, allowing seamless transition during actual relocation. The system keeps the relocation process continuous and ready-to-execute, eliminating interruptions and delays.
3Device complexity
If reactive application relocation is used, then the system is simpler to implement, but the delays in detection and processing affect QoS
Solution Approach 1:
The patent introduces preliminary action mechanisms that predict target MEC platforms and pre-establish application instances based on UE movement patterns. This adds some complexity to the relocation mechanism, but it dramatically improves QoS by eliminating detection and processing delays, ensuring continuous and reliable service during UE mobility.
Data Source
Figure 1~3
Figure 4~5
Figure 6~7
AI summary
The present invention provides an edge computing-based application relocation method and device. Said method comprises: predicting a target edge computing host for application relocation; when the target edge computing host satisfies a preset condition, making preparations on the target edge computing host, and initiating a real application relocation request to the target edge computing host, and completing application relocation at the target edge host. In the edge computing-based application relocation method according to the present invention, a target edge computing host for application relocation is predicted in advance, then preparations can be made on the target edge computing host in advance, the delay of application relocation on the edge computing host can be shortened, and the service continuity of an edge service, in particular a vehicle to everything service, in the moving process is improved, thereby better guaranteeing high reliability and low delay of a vehicle to everything application program in the moving process, improving the safety during vehicle operation.