File System Object Restoration from Raw Image Backups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data backup methods, such as file system backups, can be slow due to bottlenecks in network bandwidth and high overhead in processing and storing large numbers of small files, and raw image backups complicate the restoration of individual file system objects without maintaining the file system structure.
Innovation Solution
Implementing a technique that performs image-based backups of file systems to a networked storage location, temporarily storing raw image backup data on-premise and then copying it to off-premise storage, allowing for single file restore operations by retrieving blocks from either on-premise or off-premise storage without restoring the entire file system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If file system backups are performed, then data can be recovered, but backup speed is slow due to network bandwidth bottlenecks and high overhead in processing small files
Solution Approach 1:
The patent segments the backup data into fixed-size blocks, allowing the backup system to process and transfer data in manageable chunks rather than handling individual files. This block-based segmentation eliminates the overhead of processing numerous small files individually, thereby improving backup speed and reducing the time loss associated with file system backups.
2Productivity
If raw image backups are performed, then backup speed improves, but restoration of individual file system objects becomes complicated
Solution Approach 1:
The patent introduces an intermediary catalog data structure that maps file system objects to their corresponding backup blocks. This catalog acts as a mediator between the raw image backup format and the file system restoration process, allowing individual files to be efficiently located and restored without requiring complex parsing of the entire image backup, thus maintaining ease of operation.
3Reliability
If entire file system is restored from backup, then all data is recovered, but restore time is excessive when only individual files are needed
Solution Approach 1:
The patent enables extraction of only the specific backup blocks corresponding to requested file system objects by using the catalog for direct block location. This selective extraction capability allows the system to recover only the necessary data portions rather than restoring the entire file system, significantly reducing restore time when individual files or small sets of files are needed while maintaining reliability of data recovery.
Data Source
AI summary
Techniques associated with restoring a file system object are described in various implementations. In one example, a method may include receiving a request to restore a file system object, the file system object being stored in backup data produced from a raw image backup of a file system. The method may also include determining whether the backup data is accessible from an on-premise storage system, and retrieving blocks of data associated with the file system object from the backup data without restoring the file system. The data blocks may be retrieved from the on-premise storage system if the backup data is accessible from the on-premise storage system, or from the off-premise storage system if the backup data is not accessible from on-premise storage system. The method may also include causing the file system object to be restored.