File-Level Access to Block-Level Incremental Virtual Disk Backups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current backup systems face inefficiencies in accessing and managing file-level data from block-level backups of virtual disks, particularly in virtual machine environments, due to the need for full file system scans and the limitations of change tracking mechanisms like OS journals and file system filter drivers, which can be resource-intensive and vendor-specific.
Innovation Solution
An apparatus and method that includes a changed block module, mapping module, and changed file module to identify and map logical clusters to changed blocks, designating files as changed unless current attributes match previous backups, and updating the backup file index with current attributes and extents of changed files, enabling file-level access from block-level backups without full file system scans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full file system scans are performed to identify changed files for backup, then file-level backup accuracy is improved, but backup time and resource consumption increase significantly
Solution Approach 1:
The system performs preliminary actions by maintaining change tracking information (such as inode change journals, block change tracking) that records file and block modifications before the actual backup process. This preliminary tracking allows the backup system to identify changed files without performing full file system scans during backup, thus resolving the contradiction between backup accuracy and backup time.
2Measurement precision
If OS journals and file system filter drivers are used for change tracking, then file change detection capability is improved, but system complexity and vendor-specific dependencies increase
Solution Approach 1:
The patent introduces an intermediary layer (backup driver or filter driver) that sits between the file system and the backup software. This intermediary maintains change tracking information independently of the operating system's native journaling mechanisms, allowing the backup system to detect file changes without relying on OS-specific journals or complex filter driver chains, thereby reducing system complexity and vendor-specific dependencies.
3Productivity
If block-level incremental backups are performed without file-level access, then backup efficiency is improved, but ability to access and restore individual files deteriorates
Solution Approach 1:
The system segments the backup data structure into block-level storage units while maintaining file-level metadata and change tracking information separately. This segmentation allows the backup system to perform efficient block-level incremental backups while simultaneously providing file-level access capabilities through the maintained file metadata and change tracking structures, thus resolving the contradiction between backup efficiency and file-level access.
4Loss of substance
If changed block tracking is implemented to identify modified blocks, then data transfer volume is reduced, but complexity of mapping logical clusters to physical blocks increases
Solution Approach 1:
The system creates and maintains a copy of change tracking information (such as a copy of the changed blocks list or a mapping table) that records which logical clusters correspond to changed physical blocks. This copying approach allows the backup system to efficiently identify modified blocks without repeatedly performing complex mapping operations, thus reducing data transfer volume while managing mapping complexity through pre-computed cached information.
Data Source
AI summary
For accessing files from block-level backups of a virtual disk, a system, method, computer program product and apparatus are disclosed. The system includes an apparatus with a changed block module that determines changed blocks between a previous and a current block-level backup of a virtual disk. A mapping module maps logical clusters of the virtual disk to the changed blocks and identifies files corresponding to the logical clusters. A changed file module designates files corresponding to the logical clusters as changed files, unless current attributes of files for the current backup match attributes of the files in a backup file index corresponding to the previous backup. The changed file module stores the current attributes and extents for changed files within blocks of a backup storage device for updating in the backup file index. The method and computer program product perform these functions and may use the system and the apparatus.