Score-Based Node Selection for Balanced Resource Allocation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current cloud environments face challenges in efficiently managing resources due to inflexible development processes, slow scaling of microservices, and limitations in resource allocation, particularly in distributed and collaborative container platforms, which hinder the ability to balance resource allocation and ensure service mobility and availability.

Innovation Solution

A cloud management method and device that calculate an idle resource current state score for each node in a cluster by monitoring virtual resource usage, determining the optimal node for resource allocation based on this score, considering factors like CPU, memory, storage, security, and fault occurrence, to efficiently allocate resources for service execution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If Round Robin scheduling is used for resource allocation in distributed container platforms, then implementation simplicity is maintained, but resource allocation balance deteriorates

Engineering Contradiction:

Improvescheduling implementation complexityVSAvoidresource allocation efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent changes the scheduling parameter from simple sequential allocation (Round Robin) to score-based allocation considering multiple factors including idle resource ratios, security levels, and fault occurrence frequencies. This transforms the scheduling decision from a simple index-based selection to a multi-parameter optimization problem, improving resource allocation balance while maintaining manageable system complexity through structured scoring mechanisms.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If container-based microservices are used for rapid service distribution, then service deployment speed is improved, but resource scaling flexibility deteriorates

Engineering Contradiction:

Improveservice deployment speedVSAvoidresource scaling flexibility

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent introduces dynamic resource allocation that adapts to changing system conditions by continuously monitoring idle resource ratios, security levels, and fault frequencies. The scheduling scores are dynamically updated based on current system state, enabling the container platform to flexibly scale resources according to actual demand while maintaining rapid service deployment capabilities.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements feedback mechanisms by monitoring resource usage states, security levels, and fault occurrences, then using this information to adjust scheduling decisions. The idle resource ratio calculations and fault frequency tracking provide continuous feedback loops that enable adaptive resource scaling, allowing the system to respond to changing workloads while maintaining fast service distribution.

Inventive Principle:

Principle #23Feedback

3Device complexity

If services are constrained to single container platforms, then platform management simplicity is maintained, but service mobility deteriorates

Engineering Contradiction:

Improveplatform management complexityVSAvoidservice mobility

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal scheduling framework that can operate across multiple distributed container platforms. The standardized scoring mechanism considering idle resource ratios, security levels, and fault frequencies enables services to be scheduled across different platforms with consistent criteria, enhancing service mobility while maintaining manageable platform complexity through unified scheduling policies.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Reliability

If geographic distribution of container platforms is increased for service availability, then service availability is improved, but resource allocation coordination complexity deteriorates

Engineering Contradiction:

Improveservice availabilityVSAvoidresource allocation coordination complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent manages geographic distribution complexity by transforming resource allocation into a parameter-based scoring system. Each platform's idle resource ratios, security levels, and fault frequencies are calculated and compared using standardized parameters, enabling coordinated resource allocation across geographically distributed platforms without exponential increases in coordination complexity.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11429449B2Method for fast scheduling for balanced resource allocation in distributed and collaborative container platform environment

Publication Date: 2022.08.30 KOREA ELECTRONICS TECH INST

AI summary

A cloud management method and a cloud management device are provided. The cloud management method receives a resource allocation request for a specific service, calculates an idle resource current state score regarding each resource of each node of each cluster by monitoring virtual resource usage current states of a plurality of nodes included in a plurality of clusters, and determines a node to allocate resources for executing the requested specific service, based on the calculated idle resource current state score. Accordingly, a score is given to a resource current state of an associated cluster, and a resource candidate group is selected in response to a service scheduling request, based on the score, and an optimal node is selected based on required resources necessary for the service.