Phased Configuration Changes with Fallback Context Manager
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software applications with rich configuration variability face challenges in testing configuration changes due to the need for costly and limited dedicated testing setups, which do not accurately reflect production environments and can lead to unforeseen impacts.
Innovation Solution
Implementing a phased configuration change approach with a fallback option using a Context Manager (CM) to create, deploy, and manage new configuration-variants in parallel with the existing ones, allowing for controlled switching and merging of transactional data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a dedicated testing setup is used for configuration changes, then testing can be performed, but it is costly to setup and keep operational, limited in data scope, and may miss issues that cause negative impacts to production environment
Solution Approach 1:
The patent creates a copy of the production environment by implementing multiple contexts (production context and one or more testing contexts) within the same database schema. This allows testing to be performed in a realistic environment that mirrors production, eliminating the need for separate dedicated testing infrastructure while maintaining testing reliability.
Solution Approach 2:
The database schema is designed to support multiple contexts universally, allowing the same infrastructure to serve both production and testing purposes. The context manager enables the system to switch between different contexts, making the production environment multi-functional for both operational and testing activities.
2Adaptability or versatility
If multiple configuration versions are run in parallel in production environment, then configuration testing is improved, but data consistency and switching between versions becomes complex
Solution Approach 1:
The patent segments the database schema into multiple independent contexts, where each context represents a separate configuration version. The context manager tracks which context each data object belongs to, allowing parallel configuration versions to run independently without interfering with each other, thus managing complexity through structured segmentation.
Solution Approach 2:
The context manager acts as an intermediary layer between users and the multi-context database schema. It automatically determines which context to use based on user assignment or data object context, managing the complexity of switching between configuration versions transparently without requiring users to manually handle data consistency.
3Reliability
If a new configuration is tested with a subset of users, then rollout risk is reduced, but user assignment and context switching overhead increases
Solution Approach 1:
The patent implements dynamic context assignment where users can be assigned to different contexts based on testing requirements. The context manager dynamically determines which context a user should operate in, allowing flexible phased rollouts to subset of users while automatically managing the assignment overhead through automated context determination logic.
Data Source
AI summary
In an implementation, providing phased configuration changes with fallback option includes creating, using a context manager (CM), a new context. Using the CM, deploying a new configuration-variant. Using the CM, the new context is assigned for use by users connecting to a database schema of the new context. Using the CM, a determination is made to phase out use of the new configuration-variant. The new context is cleaned up.