Distributed Database Service Removal with Savepoint Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing the removal of services from a distributed database system while ensuring data integrity, consistency, and recoverability can be inefficient and incomplete, as existing methods may not accurately capture the removal of services post-startup, leading to potential data loss and increased costs.
Innovation Solution
A process is implemented to safely and efficiently remove a service from a distributed database system by redistributing data to remaining services, saving unsaved log segments, and registering the removal action with the database system's components, ensuring that the database can recover to a state after the service was removed, without risking data loss or inconsistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a service is stopped or shut down in prior system configurations, then the service can be removed from the system, but the database system cannot fully recover or restore to an initial state that included the removed service
Solution Approach 1:
The patent applies preliminary action by creating a savepoint (snapshot) of the database state before removing the service. This savepoint captures the configuration including the service that will be removed, enabling the database to restore to this state if needed. The savepoint is created in advance of the service removal, allowing the system to maintain recoverability while proceeding with the removal operation.
Solution Approach 2:
The patent uses copying by creating a snapshot (copy) of the database state at a specific point in time before service removal. This snapshot serves as a recoverable copy that preserves the service configuration. The system can restore from this copy if the service needs to be reinstated, thus maintaining recoverability while allowing the service to be removed from the active system.
2Adaptability or versatility
If multiple services are managed in a distributed database system, then the system can provide multiple functions and capabilities, but management of the services becomes unwieldy and costly
Solution Approach 1:
The patent applies the extraction principle by enabling the removal of individual services from the distributed database system. This allows administrators to extract and remove specific services that are no longer needed, reducing management complexity and costs while retaining the core database functionality. The service removal capability lets the system shed unnecessary components without affecting the remaining services.
Solution Approach 2:
The patent implements dynamics by making the service configuration mutable and adaptable over time. Services can be dynamically added or removed from the distributed database system based on changing requirements. The savepoint mechanism supports this dynamic behavior by capturing state at different points, allowing the system to transition between different service configurations while maintaining the ability to revert if needed.
3Productivity
If a service is removed after a last savepoint in time, then the service can be eliminated from the system, but the removal is not accurately captured or reflected in the recovery mechanism
Solution Approach 1:
The patent applies preliminary action by creating a savepoint immediately before the service removal operation. This ensures that the removal action is captured in the savepoint metadata, which is then used by the recovery mechanism. The savepoint is created in advance of the actual removal, allowing the system to accurately track and reflect the removal in future recovery operations.
Solution Approach 2:
The patent implements feedback by having the savepoint mechanism record and report the service removal action. The savepoint contains information about the service configuration at the time of creation, including the state of services. When a service is removed, this information is captured in the savepoint and fed back to the recovery mechanism, ensuring accurate tracking and reflection of the removal action.
Data Source
AI summary
A method, medium, and system to receive a request to remove a service of an instance of a distributed database system; redistribute data of the service to be removed to at least one remaining service of the distributed database system; save unsaved log segments of the service to be removed in a log backup; register removal of the service to be removed with the distributed database system; delete an entry of the service to be removed from a startup configuration of the distributed database system; halt operation of the service to be removed; and remove the halted service from internal topology information of the distributed database system.