Token-Based Data Restoration for Backup Time Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face inefficiencies in resource utilization and scalability during backup and restoration operations, particularly in reducing backup and recovery time, especially when dealing with large datasets and multiple hosts.
Innovation Solution
A method and system for efficient data restoration that utilizes a list of tokens representing a data segment's state, scanning for differences between current and previous data states, and selectively restoring only changed data portions, allowing for parallel processing and local operation without server interaction, using a bitmap to track changes and facilitate data recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional backup and restoration methods are used, then all data portions must be restored even if unchanged, but this increases backup and recovery time and consumes excessive network bandwidth
Solution Approach 1:
The data segment is divided into multiple data portions, each assigned a unique token. During restoration, the system scans each data portion and compares its token against the token from the backup state, enabling selective restoration of only changed portions rather than restoring the entire data segment.
Solution Approach 2:
The invention extracts only the changed data portions from the backup for restoration by comparing tokens. The bitmap structure extracts information about which data portions have changed, allowing the system to take out and restore only the necessary data portions rather than all data.
2Reliability
If full data restoration is performed, then data completeness is ensured, but network bandwidth consumption increases and resource utilization decreases
Solution Approach 1:
The system uses token comparison as a feedback mechanism to determine which data portions have changed. By scanning data portions and comparing their current tokens against backup tokens, the system receives feedback about what needs to be restored, enabling selective restoration that maintains data completeness while reducing network bandwidth consumption.
Solution Approach 2:
The client system performs the scanning and token comparison operations locally without requiring continuous server interaction. The system serves itself by autonomously determining which data portions need restoration through local bitmap and token comparisons, reducing the load on network resources.
3Adaptability or versatility
If centralized server-based restoration is used, then control is maintained, but system scalability is limited and resource utilization is inefficient
Solution Approach 1:
The system performs preliminary actions by maintaining bitmaps and tokens locally at the client system. During restoration operations, the client can independently scan data portions and determine what needs restoration without requiring real-time server intervention, enabling parallel processing and improving system scalability.
Solution Approach 2:
The bitmap structure serves multiple functions: it tracks changed data portions, guides the restoration process, and enables parallel scanning operations. This multi-functional approach simplifies the system architecture while improving scalability, as the same local structures support both backup and restoration operations.
Data Source
AI summary
Described are techniques for performing restoration processing for a data segment. A list of tokens is received. The list represents a first data state of said data segment. Each of the tokens correspond to a data portion of said data segment. Data portions of the data segment are scanned. A token is determined for each data portion. The token represents the data currently stored in said data segment at said each data portion. For each data portion of said data segment, it is determined whether a first token from said list associated with said each data portion is different from a second token associated with said each data portion. The second token is determined as part of said scanning and corresponds to data currently stored at said each data portion. Each data portion of the data segment determined to have a first token that is different from a second token representing data currently stored at said each data portion is restored.