Score-Based Load Balancing in Container Clusters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Managing and securing diverse applications across multiple clouds, including private and public clouds, is operationally and technologically complex due to traditional network architecture limitations, making it difficult to implement unified policies and efficiently handle workload migration and rebalancing.

Innovation Solution

A method for assigning new load to an application instance in a public cloud by calculating a resource utilization score for each instance, considering different weights for various resource metrics, and redistributing load across existing HTTP/2 server instances to optimize resource utilization without unnecessary scaling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional network architecture is used across multiple clouds, then each cloud can be managed independently with native tools, but operational complexity and technological complexity increase significantly

Engineering Contradiction:

ImproveAbility to manage diverse clouds independentlyVSAvoidOperational and technological complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges multiple cloud management operations into a unified control plane that orchestrates worker nodes across different cloud providers. This consolidation reduces operational complexity by providing centralized management while maintaining the ability to adapt to diverse cloud environments through standardized interfaces and policies.

Inventive Principle:

Principle #5Merging (Combining)

2Productivity

If new application instances are provisioned to handle increased load, then service capacity increases, but runtime costs increase due to unnecessary resource provisioning

Engineering Contradiction:

ImproveService capacityVSAvoidRuntime costs

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent implements dynamic load balancing that continuously monitors the capacity utilization of existing application instances and redistributes incoming requests accordingly. This dynamic approach allows the system to maximize the use of existing resources before provisioning new instances, thereby maintaining service capacity while minimizing runtime costs associated with unnecessary resource provisioning.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms that monitor the actual load and capacity utilization of application instances in real-time. This feedback information is used to make intelligent decisions about load distribution and instance provisioning, ensuring that new resources are only created when absolutely necessary, thus optimizing the balance between service capacity and runtime costs.

Inventive Principle:

Principle #23Feedback

3Loss of energy

If load is redistributed across existing HTTP/2 server instances, then resource utilization efficiency improves, but system complexity increases due to score calculation and rebalancing logic

Engineering Contradiction:

ImproveResource utilization efficiencyVSAvoidRebalancing system complexity

Core Design Contradiction:

Loss of energyVSDevice complexity

Solution Approach 1:

The patent employs parameter changes by calculating a resource utilization score based on multiple metrics (CPU usage, memory usage, network I/O, etc.) for each application instance. This scoring system transforms complex multi-dimensional resource states into a single comparable metric, simplifying the decision-making process for load distribution while improving resource utilization efficiency through quantitative analysis.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250130864A1Score based method to determine application load for auto-scaling and rebalancing in container-based clusters

Publication Date: 2025.04.24 VMWARE INC

AI summary

The disclosure provides a method for assigning new load to an application instance in a public cloud. The method generally includes calculating, for each application instance of a plurality of application instances running in the public cloud, a respective resource utilization score, wherein for each application instance: the respective score is calculated by applying, for each of two or more resource utilization metrics associated with the application instance, a respective weight to a respective resource usage value for the resource utilization metric, and wherein, for each of the two or more resource utilization metrics, the respective weight is a function of the respective resource usage values for the two or more resource utilization metrics; identifying an application instance having a highest respective score among the respective scores calculated for the application instances; and determining whether the application instance having the highest respective score is capable of handling the new load.