Hash-Optimized Backup System for Redundant Data Elimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data backup methods, such as incremental and differential backups, face inefficiencies when dealing with multiple computers in a network, particularly due to the limitations of the archive flag system and hash-value comparison methods, which result in redundant data backup and increased processing time.
Innovation Solution
A substantially collision-free hash-optimized backup process that generates probabilistically unique digital fingerprints for data blocks, allowing for efficient comparison with a database of stored fingerprints to determine if data blocks need to be backed up, eliminating the need for actual data comparison upon hash-value matches and reducing redundant backups across multiple computers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If incremental or differential backup is performed using archive flags, then backup time is reduced, but redundant data is backed up across multiple computers
Solution Approach 1:
A centralized backup server acts as an intermediary between multiple client computers. The server maintains a database of file identifiers and compares incoming file identifiers against this database to determine redundancy, eliminating the need for each computer to independently backup identical files.
Solution Approach 2:
The backup system implements a universal identification mechanism that works across multiple computers and file systems. By using standardized file identifiers (name, path, size, modification date) that can be universally compared, the system efficiently identifies duplicate files across different sources.
2Reliability
If full backup is performed, then data protection is comprehensive, but backup time and storage space are excessive
Solution Approach 1:
The system performs partial backups by selectively identifying and backing up only unique files that have changed since the last backup. File selection is based on comparing file identifiers against a database of previously backed-up files, implementing partial action rather than comprehensive full backup.
Solution Approach 2:
The system performs preliminary comparison of file identifiers against the backup database before executing the backup operation. This preliminary action identifies which files need to be backed up, allowing the system to prepare and execute only the necessary backup tasks.
3Reliability
If identical files on multiple computers are backed up independently, then each computer's data is protected, but storage space and processing effort are duplicated
Solution Approach 1:
The centralized backup server mediates between multiple client computers, receiving file identifier information and determining redundancy centrally. This intermediary approach allows the system to identify and eliminate duplicate files across the network, reducing total storage requirements while maintaining protection for all computers.
Solution Approach 2:
The system merges the backup operations of multiple computers into a unified process. By combining file identifier comparisons and maintaining a single centralized database of backed-up files, the system efficiently handles multiple sources without duplicating storage for identical files.
Data Source
AI summary
A method is provided to facilitate the detection of file tampering, such as a computer virus, on a computer. In one example, a digital fingerprint is generated for each file on the computer using a substantially collision-free algorithm. The digital fingerprints of the computer files are compared with digital fingerprints of the computer files generated when the files were previously saved. If the digital fingerprint of that file differs from the digital fingerprint generated when that file was previously saved, a computer virus or other tampering may exist on the file.