Backup Catalog Merging Multiple File Names to Single Inode
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current backup systems face challenges in effectively backing up and restoring files that are referenced by multiple file names, as they often result in multiple instances of the same file being stored and restored, which is inefficient and unnecessary.
Innovation Solution
A method and system that create a backup directory entry for each file to be backed up, including its unique identifier and attributes, and all names referencing it, allowing for the backup and restoration of files with multiple names while ensuring only one instance of the file is stored and restored, using APIs to determine and manage these names.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If backup systems store multiple instances of files with multiple names, then all file references are preserved, but storage space is wasted and redundancy increases
Solution Approach 1:
The patent merges multiple file references pointing to the same physical file into a single backup instance. The backup system identifies that multiple file names (hard links, symbolic links) point to the same inode and stores only one copy, thereby eliminating redundancy while preserving all file references through the backup catalog that maps each file name to its inode identifier.
Solution Approach 2:
The patent uses a logical copy mechanism where the backup catalog stores metadata references (inode mappings) instead of physical copies of the same file data. This allows the system to maintain multiple file name references without creating multiple physical instances, achieving space-efficient backup through intelligent referencing.
2Reliability
If backup systems restore multiple instances of files with multiple names, then all file references are restored, but restoration time and processing resources are wasted
Solution Approach 1:
During restoration, the patent merges multiple file reference restorations into a single file restore operation. The system queries the backup catalog to identify that multiple file names reference the same inode, then restores only one instance and recreates all file name references through the catalog information, significantly reducing restoration time and processing overhead.
Solution Approach 2:
The patent performs preliminary analysis of the backup catalog before restoration to identify all file name references to each inode. This preliminary action allows the system to plan a single restore operation that satisfies multiple file reference requirements, avoiding redundant restore operations and optimizing the restoration process.
3Loss of information
If backup systems create separate backup entries for each file name, then all file references are tracked, but backup complexity and processing overhead increase
Solution Approach 1:
The patent introduces an intermediary backup catalog that acts as a mapping layer between file names and inodes. Instead of creating separate backup entries for each file name, the system uses the catalog to store a single mapping entry per inode that lists all file names referencing it. This intermediary structure simplifies backup processing while maintaining complete file reference tracking.
Solution Approach 2:
The backup catalog serves multiple functions: it tracks all file name references, identifies duplicate files across the filesystem, manages backup operations, and supports restoration. This multi-functional intermediary structure reduces overall system complexity by consolidating multiple tracking and management tasks into a single unified mechanism.
Data Source
AI summary
Provided are a method, system, and article of manufacture for backing-up and restoring files including files referenced with multiple file names. Files to backup are determined. A determination is made from the files to backup files referenced with multiple names. An entry is created for each file to backup in a backup directory indicating file attributes and each name used to reference the file. The determined files and the backup directory are backed-up.