MEC Orchestrator for Real-Time Virtual Network Embedding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtual network embedding (VNE) methods are ineffective in Multi-access Edge Computing (MEC) networks, which require real-time resource orchestration across distributed edge areas, and are limited by propagation delays and inefficient resource allocation.
Innovation Solution
The proposed solution involves an orchestrator that calculates and implements VNE solutions in a Multi-access Edge Computing network framework, utilizing control nodes to manage resources and dynamically switch virtual nodes between standard and high-speed cores based on priority and latency requirements, optimizing resource usage and reducing processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional offline VNE methods are used, then resource allocation can be planned in advance, but they cannot handle real-time resource orchestration requirements in distributed MEC networks
Solution Approach 1:
The patent implements dynamic VNE by allowing the embedding solution to be recalculated and updated in real-time as network conditions change. The system continuously monitors resource availability and demand across distributed MEC servers, enabling adaptive resource allocation that responds to temporal variations rather than relying on static offline plans.
Solution Approach 2:
The patent divides the VNE problem into distributed segments by assigning control nodes to manage specific regions or clusters of MEC servers. This segmentation allows local decision-making to reduce propagation delays, as each control node can independently optimize resource allocation within its管辖区域 without waiting for centralized processing.
2Productivity
If resources are statically allocated in MEC networks, then allocation simplicity is maintained, but resource allocation efficiency deteriorates due to inability to adapt to changing demands
Solution Approach 1:
The patent implements feedback mechanisms where control nodes continuously monitor resource utilization metrics from MEC servers and use this information to adjust embedding solutions. This closed-loop feedback enables the system to detect resource shortages or surplus conditions and dynamically reallocate resources to optimize overall allocation efficiency.
Solution Approach 2:
The patent employs preliminary actions by pre-configuring multiple possible embedding solutions and maintaining updated resource availability information in control nodes. When resource changes occur, the system can quickly switch between pre-prepared solutions rather than calculating from scratch, reducing the complexity burden during dynamic adjustments.
3Loss of time
If virtual nodes are assigned to fixed cores, then mapping simplicity is maintained, but processing time increases due to inability to switch between standard and high-speed cores
Solution Approach 1:
The patent makes virtual node-to-core mappings dynamic by enabling control nodes to migrate virtual nodes between standard and high-speed cores based on real-time priority and latency requirements. This dynamic remapping allows the system to optimize processing performance by assigning critical tasks to high-speed cores when needed while maintaining simplicity through automated decision-making.
Solution Approach 2:
The patent changes the parameter of core speed assignment by allowing virtual nodes to be dynamically allocated to different speed tiers (standard or high-speed) based on service level requirements. This parameter flexibility enables optimization of processing time for latency-sensitive tasks without permanently configuring all nodes to high-speed, thus managing complexity through selective high-performance resource allocation.
Data Source
AI summary
A method for orchestrating resources in a multi-access edge computing (MEC) network is applied in and by an apparatus. The MEC network comprises at least one control node, substrate nodes and substrate links managed by the at least one control node. The apparatus receives a virtual network request and calculates whether a proper virtual network embedding solution for the virtual network request exists. If so, the apparatus hands the solution over to the at least one control node for implementation.