Resource-Aware MEC Server Selection for Edge Computing Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing services in mobile communication networks fail to efficiently select optimal MEC servers due to delays in real-time computing situations and congestion issues, as they primarily rely on proximity-based selection without considering service status and resource availability.
Innovation Solution
An infrastructure information apparatus that comprehensively considers MEC server locations, service status, and resource availability by interworking with cloud and Internet infrastructures to select and reselect optimal MEC servers based on user terminal movements and service requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MEC server selection is based solely on proximity to user access location, then connection setup is simple and fast, but service quality deteriorates when MEC servers become congested and cannot reflect real-time computing situations
Solution Approach 1:
The patent changes the selection parameters from solely proximity-based to a multi-parameter evaluation system that includes proximity, service status, resource availability, and real-time computing situations. The analyzer evaluates multiple parameters simultaneously to select optimal MEC servers, resolving the contradiction between simple selection and reliable service quality.
Solution Approach 2:
The system implements feedback mechanisms where the analyzer continuously monitors service status and resource status of MEC servers, and this real-time information feeds back into the selection process. This allows the system to adapt to changing conditions and maintain optimal service quality while improving selection efficiency.
2Reliability
If PDU session switching is performed to connect to optimal MEC server, then service quality improves, but processing time increases due to required reconnection procedures
Solution Approach 1:
The system performs preliminary actions by pre-evaluating and pre-selecting optimal MEC servers based on multiple parameters before actual service initiation. The analyzer proactively identifies suitable MEC servers and establishes connections in advance, so when service requests arrive, the system can quickly route to pre-selected servers without requiring time-consuming reconnection procedures.
3Device complexity
If only the closest MEC server is selected, then device complexity is minimized, but system adaptability decreases when users move or when multiple suitable servers are available
Solution Approach 1:
The patent implements a universal selection mechanism that can handle multiple selection criteria and scenarios through a single multi-parameter evaluation system. The analyzer is designed to evaluate proximity, service status, resource availability, and other factors simultaneously, making the system adaptable to various situations including user movement, server congestion, and different service requirements without increasing complexity.
Data Source
AI summary
An edge computing service-providing method and an apparatus for the same are disclosed. An infrastructure information apparatus for an edge computing service located in a control plane of a mobile communication network comprises an API server that receives a user plane function (UPF) selection request message from a session management function (SMF) of the mobile communication network when there is an edge computing service request from a user equipment (UE), a local database that stores at least one of a plurality of service names, a plurality of service identifiers corresponding to the plurality of service names, a plurality of MEC (multi-access edge computing) server identifiers, resource status information for each of the plurality of MEC servers, a UPF identifier for connection to each of the plurality of MEC servers, and a user terminal identifier connected to each of the plurality of MEC servers, and an analyzer that selects an optimal MEC server for providing the requested edge computing service through information stored in the local database or through query and response with an information system in a cloud computing infrastructure or Internet infrastructure, wherein the API server transmits a response message including a UPF identifier for connection to the selected optimal MEC server to the SMF.