AI-Driven Database Filegroup Reorganization for Space-Constrained Restores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current database restoration methods fail to accommodate drives with insufficient space in the target environment, even when the cumulative free space is larger than the cumulative data of the source database, leading to aborted restorations and inefficient space utilization.
Innovation Solution
The use of an AI engine to dynamically determine the best method for restoring the source database by reshuffling, reorganizing, and shrinking filegroups and tables within the source database to fit within the target environment's space constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the restoration process uses traditional methods that check individual drive space, then the restoration can be aborted when a single drive has insufficient space, but this prevents successful restoration even when the cumulative free space of all target drives is larger than the cumulative data size of source drives
Solution Approach 1:
The patent segments the restoration process into multiple stages: first calculating cumulative space requirements across all drives, then distributing data to individual drives based on their available space. This segmentation allows the system to handle the overall restoration as a unified problem rather than failing on individual drive constraints, enabling successful restoration when cumulative space is sufficient even if individual drives have limitations.
Solution Approach 2:
The patent performs preliminary calculations of cumulative free space and data sizes before initiating the restoration process. By pre-determining the total space requirements and available capacity, the system can plan the restoration strategy in advance and distribute data accordingly, avoiding premature termination and improving overall restoration success.
2Ease of operation
If the restoration process restores drives without reshuffling or reorganizing tables and filegroups, then the process is simple and fast, but it cannot accommodate cases where individual drives in the target environment have insufficient space for corresponding source drives
Solution Approach 1:
The patent introduces dynamic adaptability by allowing the restoration process to adjust its behavior based on the specific space constraints of the target environment. The system dynamically determines whether to perform simple direct restoration or to engage in reshuffling and reorganizing operations, making the restoration process flexible rather than rigid and accommodating various space constraint scenarios.
Solution Approach 2:
The patent changes the parameters of the restoration process by introducing options for reshuffling drives and reorganizing tables and filegroups. These parameter changes enable the system to adapt to different space configurations in the target environment, transforming the restoration from a static process into one that can modify its own behavior based on available space.
3Productivity
If the restoration process does not reshuffle drives to fit within target environment drives, then the process remains straightforward, but it fails to utilize the cumulative free space of multiple drives effectively
Solution Approach 1:
The patent merges the restoration operations across multiple drives by first aggregating the cumulative free space and then distributing data accordingly. This merging approach allows the system to treat the target environment as a unified space resource rather than independent drive units, improving space utilization efficiency while maintaining restoration speed through optimized data distribution.
Data Source
AI summary
Methods and systems are provided for restoring a source database to a target environment. The methods and systems may include calculating an amount of data on each drive of the source database. The methods and systems may include calculating free space of each drive of the target environment. The methods and systems may include shrinking filegroups in the drives of the source database. The methods and systems may include fetching historical transactional queries run on tables in the filegroups. The methods and systems may include identifying interdependencies between tables using the queries. The methods and systems may include ingesting, by an AI engine, the calculations, the queries, the identified interdependent tables, historical information of table movement and a business use case. The methods and systems may include the AI engine determining how to restore the source database using the ingested information.