MEC Resource Scheduling for Latency and Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-access Edge Computing (MEC) networks face challenges in managing varying workloads across different geographic regions due to cycles of human activity and other factors, leading to potential service disruptions and latency issues, as existing resource scheduling methods are not efficient in dynamically reallocating resources to meet changing demands.
Innovation Solution
An intelligent MEC resource scheduling service that uses distributed intelligence to monitor and predict resource usage across MEC clusters, allowing for the offloading of low-priority workloads from heavily loaded areas to lightly loaded ones, and implementing emergency resource sharing protocols to maintain quality of service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MEC resources are statically allocated at each geographic region, then local service coverage is ensured, but resource utilization efficiency deteriorates due to workload variations
Solution Approach 1:
The patent implements dynamic resource allocation by enabling MEC resources to be flexibly reallocated across different geographic regions based on real-time workload conditions. The system monitors workload metrics and automatically adjusts resource distribution, allowing resources to move from low-utilization regions to high-utilization regions, thus resolving the contradiction between maintaining service coverage and improving resource utilization efficiency
Solution Approach 2:
The patent creates a universal MEC resource pool that can serve multiple geographic regions rather than dedicating resources to single locations. The system allows any MEC resource to potentially serve any region needing service, making the resource allocation system multi-functional and adaptable to varying workload demands across different locations
2Loss of time
If MEC resources are distributed across multiple geographic regions, then local latency is reduced, but workload management complexity increases
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors workload conditions across multiple MEC regions and uses this information to make intelligent scheduling decisions. The system collects real-time data on resource utilization, service demands, and performance metrics, then automatically adjusts resource allocation to maintain low latency while simplifying management through automated control
Solution Approach 2:
The patent introduces an intelligent scheduling system that acts as an intermediary between multiple MEC regions and their workloads. This central coordinating entity manages the complexity of distributing resources across regions by making centralized scheduling decisions, thereby reducing latency through optimized resource placement while keeping management complexity contained within the scheduling system
3Device complexity
If existing resource scheduling methods are used, then system simplicity is maintained, but service disruption risk increases during workload fluctuations
Solution Approach 1:
The patent implements predictive resource allocation by analyzing historical workload patterns and predicting future resource needs before actual demand occurs. The system proactively reallocates resources in anticipation of workload fluctuations, preventing service disruptions before they happen while maintaining relatively simple system architecture through rule-based predictive algorithms
Data Source
AI summary
Systems and methods described herein provide an intelligent MEC resource scheduling service. A network device in a MEC network stores, in a memory, threshold values indicating overload conditions for resource usage by a first MEC cluster; monitors resource usage in the first MEC cluster; determines, based on the monitoring, when one of the threshold values is reached; identifies available resources in a second MEC cluster; and re-directs, based on the identifying, at least some of the resource usage from the first MEC cluster to the second MEC cluster.