Snapshot Deduplication via Reference Pointers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional backup operations redundantly send and store common data blocks across multiple snapshots, leading to inefficient use of storage space and increased network traffic.
Innovation Solution
Implementing a deduplication process that uses deduplication data to identify and represent duplicate data blocks, sending only a lightweight reference instead of the actual data, and storing only unique data blocks and their metadata across snapshots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional backup operations store all data blocks from multiple snapshots, then complete backup coverage is achieved, but storage space efficiency deteriorates due to redundant data
Solution Approach 1:
The patent creates a lightweight copy (reference pointer) instead of copying actual data blocks. When a data block is identified as duplicate across snapshots, only a reference pointer storing the original block's location is saved, rather than duplicating the entire data block content. This maintains backup completeness while dramatically reducing storage consumption.
Solution Approach 2:
The patent extracts and removes redundant data blocks from the backup storage process. By identifying data blocks that have already been backed up in previous snapshots and excluding them from subsequent snapshot backups, the system eliminates unnecessary duplication while preserving all unique data across the backup set.
2Reliability
If conventional backup operations transmit all data blocks across the network, then complete data capture is achieved, but network traffic efficiency deteriorates
Solution Approach 1:
The patent performs preliminary deduplication identification before data transmission. By pre-analyzing data blocks and identifying duplicates using hashing or comparison techniques before the backup process, the system determines which blocks need transmission and which can be referenced, thereby reducing network traffic before it occurs.
Solution Approach 2:
Instead of transmitting duplicate data blocks across the network, the patent transmits only lightweight reference pointers that indicate where the original data block is stored. This reference copying mechanism maintains data capture completeness while minimizing network bandwidth consumption.
3Volume of stationary object
If deduplication is implemented across snapshots, then storage efficiency is improved, but system complexity increases due to deduplication data management
Solution Approach 1:
The patent introduces deduplication data structures (such as hash tables or index structures) as intermediary components that facilitate efficient duplicate detection and reference management. These intermediary structures organize deduplication information in a manageable way, allowing the system to track which data blocks have been seen and where they are stored without excessive complexity.
Data Source
AI summary
Deduplicating snapshot associated with a backup operation is disclosed, including: performing a backup operation including by generating a plurality of snapshots; maintaining, at a source system, deduplication data corresponding to one or more data blocks that have already been written to backup media during the backup operation; and using the deduplication data to deduplicate backup data across the plurality of snapshots.