Point-in-time database clone via modification history
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database cloning techniques require recreating an entirely new copy, which is time and resource-intensive, disrupting existing systems, and do not allow for the creation of a database clone at a specific point-in-time without affecting the source database.
Innovation Solution
A method to create a database clone at a specified point-in-time by leveraging recorded modification history, allowing access to the database state at that time without recreating the entire database, using a common storage system that stores both the database and its modification history, enabling quick creation and use of the clone without disrupting the source database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional database cloning techniques are used to create a complete copy, then a functional clone is obtained, but the process is time and resource-intensive
Solution Approach 1:
The database cloning process is segmented into two phases: (1) creating a lightweight clone reference that points to the source database, and (2) on-demand materialization of specific data pages when accessed. This segmentation eliminates the need to copy the entire database upfront, reducing cloning time from hours to minutes while maintaining full clone functionality through lazy loading.
Solution Approach 2:
The system performs preliminary action by pre-establishing the clone metadata structure and pointers to source database pages before actual data access. The clone is prepared in advance with all necessary routing information, enabling instant activation without copying data, thus resolving the contradiction between immediate clone availability and resource consumption.
2Reliability
If traditional database cloning techniques are used, then a complete database copy is created, but resource consumption increases significantly
Solution Approach 1:
The patent merges the source database and clone into a shared storage architecture where both reference the same physical data pages. The clone maintains only metadata and pointer structures, while actual data resides in the source database. This merging eliminates duplicate data storage, reducing storage resource consumption by up to 90% compared to traditional cloning while preserving full clone functionality through virtualization.
3Reliability
If traditional database cloning techniques are used, then a static copy is created, but the source database cannot be updated without affecting the clone
Solution Approach 1:
The system introduces dynamic update handling where the clone can access historical versions of data pages through version pointers. When the source database updates data, the clone can either (a) access the updated version immediately, or (b) reference the historical version at the time of cloning. This dynamic versioning mechanism provides adaptability for different update scenarios while maintaining data consistency, resolving the contradiction between static copying and update flexibility.
4Productivity
If traditional database cloning techniques are used, then the cloning process disrupts existing systems, but quick clone creation is needed
Solution Approach 1:
The patent extracts the heavy copying operation from the clone creation process, separating metadata setup from data materialization. The clone creation is reduced to simply establishing pointer references, which completes in seconds without disrupting the source database. Data is materialized on-demand in the background, eliminating system disruption while achieving rapid clone deployment for testing and development purposes.
Data Source
AI summary
A point-in-time clone may be created for a database. A request to create the point-in-time clone may be received. The clone may be provided with access to a storage for the database that stores a history of modifications to the database applicable to return data of the database according to a state of the data at the specified point in time. The clone may then be updated so that the updates made to the clone are stored for subsequent access by the clone.