MEC Service Allocation via Local Cost Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-access edge computing (MEC) systems face variability in service improvements for user equipment (UE) due to the location of the service provider, with existing technologies struggling to optimize service allocation based on cost metrics like latency, financial cost, and quality of service (QoS), leading to suboptimal service delivery.
Innovation Solution
An MEC system incorporating a communication interface, local cost measurements module, and service allocation module that collects and analyzes local cost measurements from various entities along the path from UE to service provider, constructing a cost graph to determine the most cost-effective allocation of services to MEC hosts or service providers based on predefined policies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If services are allocated to centralized cloud computing environments, then service provider location flexibility is improved, but service latency and cost increase
Solution Approach 1:
The patent segments the centralized cloud computing environment into distributed edge computing nodes (MEC hosts) located at different network edges. This segmentation allows services to be allocated to geographically distributed locations closer to UEs, reducing latency while maintaining provider flexibility through the cost graph evaluation framework that compares multiple edge location options.
Solution Approach 2:
The patent introduces a new dimensional approach by creating a cost graph that evaluates service allocation across multiple dimensions simultaneously (latency, financial cost, QoS metrics). This multi-dimensional evaluation enables optimal service placement at edge locations that balance proximity to UEs with overall service quality requirements.
2Productivity
If services are allocated based on multiple cost metrics (latency, financial cost, QoS), then service optimization is improved, but system complexity increases
Solution Approach 1:
The patent introduces a MEC platform manager as an intermediary component that centralizes the complex task of multi-metric evaluation. The platform manager collects cost metrics from multiple sources, constructs the cost graph, and makes allocation decisions, thereby simplifying the overall system architecture while enabling comprehensive service optimization across latency, financial cost, and QoS dimensions.
3Measurement precision
If local cost measurements are collected from multiple entities along the service path, then allocation accuracy is improved, but measurement and data collection complexity increases
Solution Approach 1:
The patent implements a universal cost measurement framework where the MEC platform manager serves multiple functions: collecting metrics from diverse entities (UEs, access points, MEC hosts), constructing the cost graph, evaluating service allocations, and making decisions. This multi-functional approach consolidates complex measurement tasks into a single coordinating system, improving allocation accuracy while managing data collection complexity through centralized orchestration.
Data Source
AI summary
Embodiments herein may include systems, apparatuses, methods, and computer-readable media, for a multi-access edge computing (MEC) system. An apparatus for MEC may include a communication interface, a local cost measurements module, and a service allocation module. The communication interface may receive, from a UE, a request for a service to be provided to the UE. The local cost measurements module may collect a set of local cost measurements for the service. The service allocation module may determine to allocate the service to a MEC host based on an allocation policy related to a cost for the MEC host to provide the service or a cost for a service provider to provide the service in view of the one or more local cost measurements. Other embodiments may be described and/or claimed.