Incremental Backup Metadata Generation via Block Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current block-based backup solutions incur significant CPU cycles and time during incremental backups due to the need to parse and transmit entire file system metadata, even when only a few files have changed since the last backup, leading to inefficiencies in data processing and recovery.
Innovation Solution
The solution involves noting and timestamping file system metadata blocks during a full backup, filtering and processing only changed metadata blocks during incremental backups, and using parent inode information to update metadata files, thereby reducing the need to back up and process unchanged metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If entire file system metadata is parsed and stored during incremental backup, then file-by-file recovery is enabled, but CPU cycles and time consumption increase significantly
Solution Approach 1:
The patent extracts only the changed metadata blocks from the entire file system metadata during incremental backup, rather than processing all metadata. By identifying and isolating only the modified portions (through block-level change detection and metadata block filtering), the system enables selective backup of minimal necessary data, thus maintaining recovery capability while dramatically reducing processing overhead
Solution Approach 2:
The patent segments file system metadata into individual blocks and processes them independently. By dividing the metadata into discrete blocks that can be individually identified, tracked, and processed, the system can apply change detection at the block level and only backup modified blocks, avoiding the need to parse and transmit entire metadata structures
2Reliability
If all file system metadata is transmitted during incremental backup, then complete file system state is captured, but data transmission volume and processing time increase
Solution Approach 1:
The patent applies partial action by transmitting only the subset of metadata blocks that have changed since the last backup, rather than transmitting all metadata. By performing change detection at the block level and selectively transmitting only modified blocks (plus necessary parent directory information), the system captures the complete changed state while minimizing transmission volume and processing time
3Reliability
If parent directory metadata is included in backup, then file system hierarchy is preserved, but additional processing overhead is introduced
Solution Approach 1:
The patent applies local quality by selectively including parent directory metadata information only when necessary for the changed blocks being backed up. Rather than uniformly processing all parent directories, the system identifies and processes only those parent directory blocks that are relevant to the changed files, applying different processing intensity to different parts of the file system based on local needs
Data Source
AI summary
Metadata generate for incremental backup is disclosed. A subset of blocks used to store file system metadata are identified in a set of blocks changed since a last backup. File system metadata stored in the subset of blocks is used to obtain file system metadata associated with file system objects that have been created and/or modified since the last backup. The file system metadata associated with file system objects that have been created and/or modified since the last backup is used to generate file system metadata files for the incremental backup.