MEC Resource Allocation via Mobile Terminal Course Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
MEC servers with a hierarchical structure face quantitative limitations in resource deployment, leading to reduced resource utilization efficiency due to uniform overcommitment, which can cause performance degradation and operational instability.
Innovation Solution
A resource allocation device that estimates the movement status of mobile terminals using network information to dynamically adjust resource allocation and release, ensuring the overcommit ratio does not exceed a set limit, thereby optimizing resource utilization based on predicted terminal locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If uniform overcommitment is applied to improve resource utilization efficiency, then resource utilization efficiency is improved, but performance degradation and operational instability occur
Solution Approach 1:
The patent implements dynamic overcommitment by continuously monitoring mobile terminal locations and adjusting resource allocation in real-time. The system calculates the number of mobile terminals in each area and dynamically adjusts the overcommit ratio accordingly, transitioning from static uniform overcommitment to adaptive dynamic overcommitment that responds to changing conditions.
Solution Approach 2:
The patent applies different overcommit ratios to different geographic areas based on local mobile terminal density. Instead of uniform overcommitment across all areas, the system calculates area-specific overcommit ratios by dividing the number of mobile terminals in each area by the number of physical servers in that area, allowing each region to have optimized resource allocation tailored to its specific conditions.
2Reliability
If resources are always reserved for all mobile terminals to ensure real-time processing, then real-time processing capability is ensured, but resource utilization efficiency is reduced
Solution Approach 1:
The system performs preliminary resource allocation by predicting mobile terminal locations and pre-allocating resources to MEC servers in advance. The course estimation unit predicts future locations of mobile terminals, and resources are allocated to the predicted destination MEC servers before terminals actually arrive, ensuring real-time processing capability while avoiding unnecessary resource reservation.
Solution Approach 2:
The patent implements dynamic resource reservation that adjusts to actual mobile terminal distribution. Instead of statically reserving resources for all possible terminals, the system continuously updates resource allocation based on real-time location data and movement patterns, releasing resources from areas with low terminal density and allocating to areas with high density, thus maintaining real-time processing while improving utilization.
3Productivity
If the upper limit of overcommit ratio is raised at design stage to improve accommodation rate, then accommodation rate is improved, but performance degradation occurs
Solution Approach 1:
The system dynamically adjusts the effective overcommit ratio based on real-time conditions rather than using a fixed high upper limit. The execution unit calculates the actual overcommit ratio for each area and only applies it when it does not exceed the predetermined upper limit, allowing the system to achieve high accommodation rates during low-demand periods while automatically reducing overcommitment during high-demand periods to maintain performance stability.
Data Source
AI summary
A resource allocation device includes: a course estimation unit that estimates a course of each mobile terminal based on a location of a mobile terminal acquired from a network device, and estimates a probability that each mobile terminal is located in each area at time of prediction; a determination unit that calculates, for each area, the number of mobile terminals in the area using the probability and determines whether the maximum value of an overcommit ratio for each area exceeds an upper limit; and an execution unit that executes the allocation or release of resources to an MEC server group located in each area when the maximum value of the overcommit ratio is equal to or less than the upper limit, and refrains from executing the allocation or release of the resources to the MEC server group when the maximum value of the overcommit ratio exceeds the upper limit.