Balanced Resource Scaling in Distributed Container Platforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud management systems face challenges in efficiently scaling resources to handle increasing traffic and provide flexible, geographically distributed services, as they often rely on one-dimensional scaling methods that limit resource allocation and mobility between container platforms.
Innovation Solution
A cloud management method and device that determine whether pods are overloaded, identify resource usage states, and dynamically scale resources by increasing allocations, creating pod replicas, or redistributing workload across nodes and clusters based on available resources, enabling balanced and flexible scaling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If container-based micro services are used to effectively use internal resources and distribute applications promptly, then resource utilization and deployment speed are improved, but the system has a limit to scaling out resources in response to increasing user traffic
Solution Approach 1:
The patent transitions from traditional horizontal scaling (adding more nodes) to multi-dimensional scaling that includes vertical scaling (adding resources to existing nodes), cross-node scaling (distributing workloads across nodes), and cross-cluster scaling (expanding to additional clusters). This dimensional expansion enables the system to scale resources in multiple directions simultaneously, overcoming the limitations of conventional single-dimension scaling approaches.
Solution Approach 2:
The patent segments the scaling operation into distinct dimensions: vertical scaling within a node, horizontal scaling across nodes, and cross-cluster scaling. By dividing the scaling process into these manageable segments, the system can apply different scaling strategies to different dimensions based on resource availability and workload requirements, enabling more flexible and effective resource expansion.
2Ease of manufacture
If existing cloud resource scaling-out methods provide one-dimensional scale out horizontally or vertically, then implementation simplicity is improved, but the limit to scaling out resources is increased
Solution Approach 1:
The patent extends the scaling approach from one dimension to multiple dimensions by incorporating vertical scaling, horizontal scaling, cross-node scaling, and cross-cluster scaling. This multi-dimensional approach maintains implementation simplicity through automated orchestration while dramatically improving resource scaling flexibility and adaptability to various workload scenarios.
Solution Approach 2:
The patent creates a universal scaling mechanism that can operate across multiple dimensions and contexts. The same scaling framework can perform vertical scaling within a node, horizontal scaling across nodes, or cross-cluster scaling depending on resource availability and requirements, making the system adaptable to diverse scaling scenarios without requiring separate implementations for each scaling type.
3Stability of the object's composition
If micro services do not allow mobility of services between container platforms, then platform stability is improved, but service mobility and flexibility are reduced
Solution Approach 1:
The patent employs copying mechanisms to replicate pods and services across different nodes and clusters. By creating copies of service instances and distributing them across the container platform infrastructure, the system enables service mobility while maintaining operational stability. The copying approach allows services to be relocated and scaled across platforms without disrupting the underlying platform architecture.
Data Source
AI summary
A cloud management method and a cloud management device are provided. The cloud management method determines whether a plurality of pods are overloaded, identifies resource usage current states of a cluster and a node, and determines a method of scaling resources of a specific pod that is overloaded from among the plurality of pods, according to the resource usage current states of the cluster and the node, and scales the resources of the specific pod according to the determined method. Accordingly, scaling for uniformly extending resources of a node and a pod in a cluster horizontally and vertically can be automatically performed.