Selective Restore from Block Level Backup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Block level backups lack the ability to selectively restore individual files or folders, requiring a full disk restore, which hampers widespread adoption due to the need for restoring entire disks instead of specific data.
Innovation Solution
A method to create a child disk from a block level backup, allowing for the selection and restoration of specific files and folders by populating the child disk with data from corresponding blocks, enabling file/folder and application restores, and supporting virtual machine recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If block level backup is performed, then backup efficiency and storage utilization are improved, but the ability to selectively restore individual files or folders deteriorates
Solution Approach 1:
The patent segments the backup system into two distinct components: block-level backup operations for efficient data capture, and file-level metadata indexing for selective restoration. The backup agent divides the disk into blocks and backs up only changed blocks, while separately maintaining an index that maps file systems to their corresponding block locations. This segmentation allows independent optimization of backup efficiency and restore selectivity.
Solution Approach 2:
The patent introduces an intermediary metadata index structure that acts as a bridge between block-level backup data and file-level restoration requirements. This index contains file system information, directory structures, and mappings between files and their constituent blocks, enabling the restoration software to selectively identify and restore only the blocks corresponding to desired files without manual block-by-block reconstruction.
2Reliability
If full disk restore is performed, then complete data recovery is achieved, but restoration time and resource consumption increase
Solution Approach 1:
The patent extracts only the necessary file system metadata and directory information from the complete disk image, storing it separately in an index. During restoration, this extracted metadata allows the system to identify and restore only the specific files or folders needed, rather than restoring the entire disk. This extraction principle enables partial restoration while maintaining data integrity for the selected portions.
Solution Approach 2:
The patent implements partial restoration by allowing users to select specific files, folders, or even individual blocks for restoration based on the metadata index. The restoration process performs only the necessary actions to recover the selected data portions, avoiding the excessive action of restoring entire disks when only small portions are needed, thereby significantly reducing restoration time and resource usage.
3Quantity of substance
If block level backup is used, then storage space is optimized, but understanding of files/folders/applications deteriorates
Solution Approach 1:
The patent segments information storage into two separate structures: block-level backup data for storage efficiency and file-level metadata index for information understanding. The backup agent maintains the actual data in block form while separately indexing file names, directory structures, application information, and block mappings. This segmentation preserves storage optimization while preventing information loss about the logical organization of data.
Solution Approach 2:
The patent creates a copied representation of the file system structure in the metadata index without duplicating the actual data blocks. The index contains copies of file names, folder hierarchies, application metadata, and mappings between logical file structures and physical block locations. This copying approach provides complete file system understanding while maintaining storage efficiency by referencing rather than duplicating the actual data blocks.
Data Source
AI summary
A computer implemented method includes creating a child disk for a machine. A configuration of a backed up disk drive is derived from a storage container that includes a block level backup copy of the disk drive. The child disk is populated with selected data from blocks of data corresponding to selected files desired to be used on the machine. Drivers may be injected and registries modified to start with booting of the virtual.