Dynamic Virtual Machine Configuration via Abstract Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing system configuration methods are static and redundant, requiring manual updates for each node with system-dependent information, making it difficult to adapt configurations across different hardware and software environments, leading to downtime and inefficiencies in managing Java-based systems.
Innovation Solution
An abstract configuration model that uses parameterized and computed settings, along with inheritance principles, to dynamically configure virtual machines and components, allowing for system-independent templates that adapt to specific environments and scenarios, reducing the need for manual updates and enabling flexible configuration management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static and redundant configuration distribution is used across the cluster, then each node can be individually configured, but manual updates are required for every node when system changes occur
Solution Approach 1:
The patent combines multiple individual node configurations into a single centralized configuration file that manages all nodes uniformly. This allows system administrators to update configuration parameters once, and the changes are automatically propagated to all nodes in the cluster, eliminating the need for manual updates on each node while preserving individual node configurability through the unified interface.
2Ease of operation
If system-dependent information is statically distributed, then configurations can be manually configured per node, but adaptability to different hardware and software environments is lost
Solution Approach 1:
The patent transforms static configuration values into dynamic parameters that can be adjusted based on the runtime environment. The configuration system detects hardware and software characteristics automatically and adapts settings accordingly, while still allowing manual override when needed. This enables the same configuration file to work across diverse environments without requiring manual reconfiguration for each node.
3Reliability
If VM settings changes are applied, then the configuration is updated, but the instance must be restarted causing downtime
Solution Approach 1:
The patent performs configuration validation and preparation in advance before applying changes to running instances. The system pre-processes configuration files, checks for compatibility and errors, and prepares update packages that can be applied atomically. This preliminary action ensures that when changes are deployed, they can be applied without requiring instance restarts, or if restarts are necessary, the system can perform health checks and failback procedures to minimize downtime.
4Device complexity
If Java parameters are stored in a single property with complete overwrite inheritance, then child configurations override parent parameters, but parent changes cannot be propagated to children
Solution Approach 1:
The patent implements a hierarchical parameter inheritance system where configuration parameters are structured with explicit inheritance rules. Child configurations inherit parameters from parent configurations by default, and the system provides mechanisms to selectively override specific parameters. When parent parameters change, the system automatically propagates these changes to child configurations unless explicitly marked as overridden, maintaining both the simplicity of a unified configuration structure and the flexibility of parameter propagation.
Data Source
AI summary
A deployable template structure is described. In an embodiment, the deployable template structure includes sections for virtual machines deployable, filters, and component configuration settings. Parameters for the virtual machine section are broken down into three groups: memory, system, and additional.