Remote Virtual Database Provisioning with Snapshot Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional database systems impose limitations by requiring the replication stream to be stopped for disaster recovery, preventing the use of secondary databases for updates and limiting topology configurations between primary and secondary systems.
Innovation Solution
The system allows remote provisioning of virtual databases, enabling the target database system to be used actively while replication from the source continues, by storing snapshots of data and managing database infrastructure objects to prevent storage reclamation when dependent objects exist, thus supporting continuous replication and updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the replication stream is stopped to use the target database system for updates, then the target system can be actively used, but the replication continuity is interrupted and productivity decreases
Solution Approach 1:
The system segments the database infrastructure into independent objects (source databases, snapshots, virtual databases) that can be managed separately. The target database system can create virtual databases from snapshots independently while the replication stream continues uninterrupted, allowing active usage without stopping replication.
Solution Approach 2:
Snapshots serve as intermediaries between the source database and the target virtual databases. The target system creates virtual databases from snapshots rather than directly from the source, allowing the replication stream to continue while the target system actively serves update requests.
2Reliability
If the target database system is used passively for disaster recovery only, then data integrity is maintained, but the system versatility and adaptability are limited
Solution Approach 1:
The target database system is designed to perform multiple functions: it can serve as a disaster recovery backup (passive mode) and simultaneously function as an active database system that creates and serves virtual databases. The system maintains data integrity through snapshot-based provisioning while enabling diverse topologies including active-active and active-passive configurations.
Solution Approach 2:
The system allows dynamic role assignment where the target system can switch between passive disaster recovery mode and active virtual database hosting mode. The infrastructure objects can be independently created, activated, or deactivated based on operational needs, providing flexible adaptability while maintaining data integrity.
3Loss of substance
If database blocks are reclaimed when source database objects are deleted, then storage efficiency is improved, but data availability for dependent virtual databases is compromised
Solution Approach 1:
The system implements feedback mechanisms where the target database system monitors dependencies between infrastructure objects. When a source database object is deleted, the system checks whether any virtual databases depend on it. Storage blocks are only reclaimed when no dependent virtual databases exist, ensuring data availability while optimizing storage space utilization.
Solution Approach 2:
The system performs preliminary dependency checking before reclamation. When deletion is requested, the system first verifies whether any active virtual databases depend on the infrastructure object. This preliminary action prevents premature reclamation that would compromise data availability while enabling safe reclamation when no dependencies exist.
Data Source
AI summary
Database storage systems provide replication capability that allows data of a source database storage system to be replicated to a target database storage system. Virtual databases can be provisioned from the target database system, thereby supporting remote provisioning of virtual databases. The virtual databases on the remote site can be used to perform various operations including read/write of data, refresh, rollback, and so on. Database objects may be deleted on the source database storage system while the corresponding data is used by other entities for example, virtual databases at the target database storage system. The database storage system verifies if an entity being deleted is in use by any other database infrastructure object of the database storage system. A placeholder object maintains the data corresponding to the deleted object. This provides the ability to create geographical distribution networks and support remote provisioning of virtual databases.


