MEC Server Clustering via Virtual Regions for Resource Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

MEC servers are physically and geographically distributed, leading to inefficient resource utilization and operation due to varying resource availability and physical restrictions, making it difficult to provide functions/services efficiently across different server groups with hierarchical structures.

Innovation Solution

A clustering system that monitors MEC server and network usage, generates cluster plans based on telemetry and non-telemetry information, and constructs clusters within virtual regions to optimize resource allocation and meet service requirements, considering costs and maintenance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If MEC servers are physically and geographically distributed to provide services near terminals, then service response speed and terminal proximity are improved, but resource utilization efficiency deteriorates due to varying resource availability across different locations

Engineering Contradiction:
Improveservice response speedVSAvoidresource utilization efficiency
Core Design Contradiction:
SpeedVSProductivity

Solution Approach 1:

The system segments the geographically distributed MEC servers into virtual regions based on service requirements and resource characteristics. Each virtual region can be independently managed and optimized, allowing efficient resource allocation across multiple locations while maintaining the geographical distribution benefit for service response speed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a virtual dimension (virtual regions) overlaying the physical geographical distribution. This allows resources to be organized and allocated across multiple geographical locations without being constrained by physical proximity, thereby improving resource utilization efficiency while maintaining the speed benefits of geographical distribution.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Adaptability or versatility

If MEC servers are deployed at multiple geographically different bases with hierarchical structure, then service coverage and accessibility are improved, but system complexity increases due to varying physical restrictions and resource availability

Engineering Contradiction:
Improveservice coverageVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The virtual region management system provides a universal framework that can accommodate multiple MEC servers across different geographical bases with varying physical restrictions. The system handles diverse resource types and service requirements through a unified approach, reducing the complexity of managing heterogeneous distributed systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces virtual region management as an intermediary layer between the physical MEC servers and the service provisioning system. This intermediary abstracts the complexity of geographical distribution and physical restrictions, providing a simplified interface for service deployment while maintaining comprehensive service coverage.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If cluster configuration is used for redundancy to prevent service interruption, then service reliability is improved, but resource overhead and system cost increase

Engineering Contradiction:
Improveservice reliabilityVSAvoidresource overhead
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The system merges multiple MEC servers into virtual regions to form clusters that provide redundancy and reliability. By combining resources across geographical locations, the system achieves fault tolerance and service continuity while optimizing resource utilization to minimize overhead compared to traditional local clustering approaches.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11875192B2Clustering system and clustering method for providing services using multi- access edge computing servers
Publication Date: 2024.01.16 NIPPON TELEGRAPH & TELEPHONE CORP
  • US11875192B2 patent drawing
  • US11875192B2 patent drawing
  • US11875192B2 patent drawing

AI summary

A system and method capable of efficiently using and operating resources and allowing a cluster satisfying requirements of functions/services provided to terminals to be configured of multi-access edge computing (MEC) servers are provided. A clustering system 100 includes a monitoring unit 110 configured to monitor a usage status of MEC servers and a network and store telemetry information in a storage unit 120, a cluster planning request unit 130 configured to acquire non-telemetry information such as provision area information and requirement information of functions/services and deployment information of the MEC server, store the non-telemetry information in the storage unit 120, a cluster planning unit 140 configured to set a virtual region including one or more bases satisfying a provision area and requirements on the basis of each piece of information in the storage unit 120 and calculate a cluster plan for providing the functions/services from a plurality of MEC servers disposed at the base belonging to the virtual region, and a cluster construction unit 150 configured to perform a clustering process on the basis of the cluster plan.