MEC Server Selection Orchestrator for Multi-Client Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing architectures face challenges in optimizing the selection of Multi-Access Edge Computing (MEC) servers for multi-client applications, particularly in ensuring fair and efficient performance across diverse client devices located at different geographical locations, due to varying latency and resource requirements.
Innovation Solution
A system and method for orchestrating the selection of MEC servers that dynamically account for optimization policies, such as latency minimization or latency equalization, by identifying candidate MEC servers based on performance parameters and client device locations, and selecting the optimal server to ensure fair and efficient service delivery across multiple client devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If MEC servers are distributed to multiple nodes closer to client devices, then latency is reduced and scalability is improved, but it becomes challenging to optimize server selection for multi-client applications with different goals and locations
Solution Approach 1:
The patent introduces an orchestration system as an intermediary between client devices and MEC servers. This orchestration system receives requests from client devices, evaluates multiple MEC servers based on performance parameters and optimization policies, and selects the optimal server. This mediator approach resolves the contradiction by maintaining low latency through distributed servers while simplifying the selection process through centralized intelligent orchestration.
Solution Approach 2:
The patent implements feedback mechanisms where the orchestration system continuously monitors performance parameters of MEC servers and uses this information to make informed selection decisions. The system evaluates candidate servers based on real-time performance data and optimization policies, creating a feedback loop that optimizes server selection dynamically. This resolves the complexity by using automated feedback-driven decision-making rather than static or manual selection.
2Loss of time
If MEC servers are selected based on proximity to client devices, then latency is minimized for individual clients, but fairness among multiple clients in multi-client applications deteriorates
Solution Approach 1:
The patent applies local quality by allowing different optimization policies to be applied to different multi-client applications based on their specific requirements. The orchestration system can select servers that optimize for latency for some applications while prioritizing fairness for others. This resolves the contradiction by making the selection criteria adaptive to local application needs rather than applying a single global rule.
Solution Approach 2:
The patent implements dynamic server selection where the orchestration system can change selection criteria based on current application requirements and conditions. For multi-client applications, the system dynamically evaluates candidate servers against optimization policies that may prioritize fairness over minimum latency. This dynamic approach resolves the contradiction by allowing the system to adapt between latency optimization and fairness based on application-specific goals.
3Productivity
If a single MEC server is assigned to serve all client devices in a multi-client application, then resource utilization is improved, but performance fairness among clients located at different geographical locations deteriorates
Solution Approach 1:
The patent applies segmentation by dividing the server selection process into multiple stages: identifying candidate servers, evaluating them against optimization policies, and selecting the optimal server. The system segments the evaluation criteria into performance parameters and application-specific optimization policies. This resolves the contradiction by systematically balancing resource utilization considerations with latency fairness through structured multi-criteria evaluation.
Data Source
AI summary
An exemplary multi-access edge computing (MEC) orchestration system obtains an operation parameter of a multi-client application that is to execute on a MEC server to be selected from a set of MEC servers located at a first set of locations within a coverage area of a provider network. When executing, the multi-client application serves respective client applications of a set of client devices located at a second set of locations within the coverage area. Based on the operation parameter, the MEC orchestration system identifies a candidate subset of MEC servers from the set of MEC servers and directs the set of client devices to test and report performance capabilities of the MEC servers in the candidate subset. Based on the reported performance capabilities, the MEC orchestration system selects, from the candidate subset of MEC servers, a particular MEC server on which the multi-client application is to execute.