Edge Computing Orchestrator Latency Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing a large number of Multi-access Edge Computing (MEC) devices poses challenges in selecting the optimal MEC device for user equipment (UE) devices due to varying latency requirements and hardware capabilities, as MEC devices may not always be able to service requests efficiently, especially when they lack necessary resources like AI accelerators or exceed latency thresholds.
Innovation Solution
An orchestrator system provides test configurations to MEC devices, collects performance data, and generates reports to select the most suitable MEC device based on latency and capability requirements, while also alerting administrators when thresholds are approached or exceeded, ensuring efficient service delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If MEC devices are deployed at the network edge to reduce latency, then user quality of experience is improved, but device complexity and management difficulty increase
Solution Approach 1:
The patent introduces an orchestrator system as an intermediary between the core network and distributed MEC devices. This orchestrator manages device discovery, capability assessment, and service routing, thereby reducing the management complexity burden on individual network elements while maintaining low-latency edge computing capabilities.
Solution Approach 2:
The patent implements feedback mechanisms where MEC devices report their capability status (including AI accelerator availability and latency metrics) to the orchestrator. The orchestrator uses this feedback to dynamically select appropriate MEC devices for specific services, optimizing both latency performance and management efficiency.
2Loss of time
If MEC devices are selected based on proximity to UE devices, then latency is reduced, but service reliability deteriorates when MEC devices lack necessary resources
Solution Approach 1:
The patent changes the selection parameters for MEC devices from purely geographic proximity to a multi-dimensional assessment including capability parameters (AI accelerator presence, processing power, memory) and performance parameters (latency metrics, load status). This allows the system to select the most suitable MEC device based on both location and resource availability.
Solution Approach 2:
The patent implements preliminary capability assessment and testing of MEC devices before service deployment. The orchestrator pre-evaluates MEC device capabilities and maintains an updated registry, enabling rapid and reliable service routing decisions without requiring real-time resource verification when services are requested.
3Measurement precision
If performance testing is conducted on all MEC devices, then selection accuracy is improved, but time and computational resources are consumed
Solution Approach 1:
The patent implements partial testing strategies where not all MEC devices undergo complete performance testing. Instead, the orchestrator performs selective testing based on device category, service type, and current network conditions. Basic capability assessments are performed on all devices, while detailed performance testing is conducted only on candidate devices that pass initial screening.
Data Source
AI summary
A computer device may include a memory storing instructions and processor configured to execute the instructions to provide a test configuration to a plurality of multi-access edge computing (MEC) devices; collect test results associated with the provided test configuration from the plurality of MEC devices; and obtain capability information associated with particular ones of the plurality of MEC devices. The computer device may be further configured to generate a test report that relates one or more parameters included in the test results and the obtained capability information to particular ones of the plurality of MEC devices; and use the generated test report to select a MEC device from the plurality of MEC devices for a user equipment (UE) device requesting an application session.