Dynamic Scaling via Automatic Manifest Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing dynamic scaling mechanisms in container orchestration platforms are inadequate for complex applications, requiring significant user effort to implement separate scaling mechanisms and often resulting in ineffective dynamic scaling due to repeated modifications in configuration manifests.
Innovation Solution
A method that acquires current metric values for application components, determines a target number of replicas based on a scaling policy, and updates the configuration manifest automatically, enabling dynamic scaling without additional user effort by leveraging the container orchestration system's ability to perform scaling based on the updated manifest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing simple dynamic scaling mechanism is used, then implementation is easy, but it cannot meet dynamic scaling needs of complex applications
Solution Approach 1:
The patent segments the scaling mechanism into independent component-level configurations. Each component can have its own scaling policy defined in the configuration manifest, allowing complex applications to have different scaling behaviors for different components based on their specific needs, while maintaining the simplicity of a unified configuration approach.
Solution Approach 2:
The patent enables dynamic scaling by allowing the configuration manifest to be automatically updated based on runtime metric values and scaling policies. The system continuously monitors metrics and adjusts replica numbers dynamically without requiring manual intervention, making the scaling mechanism adaptable to changing application requirements.
2Adaptability or versatility
If users implement separate dynamic scaling mechanisms manually, then scaling can be customized, but user effort and time consumption increase significantly
Solution Approach 1:
The patent implements self-service scaling by enabling the system to automatically monitor metric values, evaluate scaling policies, and update configuration manifests without user intervention. Users only need to initially configure the scaling policies in the manifest, after which the system autonomously performs all scaling operations, eliminating repeated manual efforts.
Solution Approach 2:
The patent establishes a feedback loop where the system continuously monitors runtime metric values, compares them against scaling policies, and automatically adjusts replica numbers accordingly. This closed-loop feedback mechanism ensures that scaling decisions are based on actual application performance and resource utilization, eliminating the need for manual tuning and adjustment.
3Productivity
If configuration manifest is repeatedly modified for scaling, then dynamic scaling can be achieved, but system stability decreases due to repeated modifications
Solution Approach 1:
The patent applies preliminary action by pre-defining scaling policies and thresholds in the configuration manifest before runtime. These pre-configured policies guide the automatic scaling decisions, reducing the need for frequent manifest modifications. The system only updates the manifest when scaling actions are actually required based on metric evaluations.
Solution Approach 2:
The patent implements periodic monitoring and evaluation of metric values against scaling policies. Instead of continuously modifying the configuration manifest, the system periodically checks whether scaling actions are needed and only updates the manifest when changes are warranted, thereby maintaining configuration stability while still achieving responsive dynamic scaling.
Data Source
AI summary
Support for dynamic scaling is provided. An example method includes: acquiring a current value of a metric associated with a plurality of components of an application; determining a target number of replicas of at least one component of the plurality of components based on the current value of the metric and a scaling policy for the plurality of components; and updating a configuration manifest of the application based on the target number of replicas of the at least one component. In this manner, the configuration manifest of the application can be automatically updated based on metric values of the components at runtime and a custom scaling policy, thereby achieving dynamic scaling of various components of the application without additional user effort to implement separate dynamic scaling mechanisms.