Differential Data Backup via Fingerprint Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In remote replication technologies, the process of data backup is resource-intensive due to the need for calculating and comparing fingerprints of all data blocks, leading to high consumption of computing resources and network bandwidth.

Innovation Solution

A method and device for differential data backup that determine the fingerprint information set corresponding to a current backup period, allowing only the necessary data blocks to be backed up, reducing the need for extensive fingerprint calculations and comparisons.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the source storage device sends fingerprints of all data blocks to the backup storage device, then the backup storage device can accurately determine differential data, but the consumption of computing resources and network bandwidth increases significantly

Engineering Contradiction:
Improveaccuracy of differential data determinationVSAvoidconsumption of computing resources and network bandwidth
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

The patent segments the data backup process into two phases: first, the source storage device calculates and sends fingerprints only of data blocks that have changed since the last backup; second, the backup storage device uses these fingerprints to identify and backup only the differential data. This segmentation avoids the need to process all data blocks, thereby reducing computing resource consumption and network bandwidth usage while maintaining accurate differential data determination.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The source storage device performs preliminary action by calculating and identifying which data blocks need backup before actually transmitting data to the backup storage device. The fingerprint calculation and comparison are performed in advance on the source side, allowing the backup process to focus only on necessary data blocks, thus reducing overall resource consumption.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If the source storage device calculates fingerprints of all data blocks, then all data can be identified for backup, but the computing resource consumption increases

Engineering Contradiction:
Improvequantity of data blocks identified for backupVSAvoidcomputing resource consumption
Core Design Contradiction:
Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

Instead of calculating fingerprints for all data blocks (excessive action), the patent applies partial action by calculating and sending fingerprints only for data blocks that have changed since the last backup. This partial approach identifies and backups only the necessary differential data, significantly reducing computing resource consumption while ensuring all required data is captured.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If the backup storage device compares all received fingerprints, then accurate differential data identification is achieved, but the network bandwidth consumption increases

Engineering Contradiction:
Improveaccuracy of differential fingerprint identificationVSAvoidnetwork bandwidth consumption
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent extracts and transmits only the essential information (fingerprints of changed data blocks) from the source storage device to the backup storage device, rather than transmitting all data block fingerprints. This extraction approach maintains the ability to accurately identify differential data while significantly reducing the quantity of data transmitted over the network.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3312727B1Differential data backup method and device
Publication Date: 2018.11.14 HUAWEI TECH CO LTD
  • EP3312727B1 patent drawingFigure 1
  • EP3312727B1 patent drawingFigure 2
  • EP3312727B1 patent drawingFigure 3~4

AI summary

The present invention provides a method and a device for differential data backup. The method is applied to a storage system, the storage system includes a source storage device and a backup storage device, and the method is executed by the source storage device. The method includes: determining, according to an identifier of a current backup period and a correspondence between an identifier of a backup period and a fingerprint information set, a fingerprint information set corresponding to the current backup period, where the fingerprint information set includes fingerprint information of a target data block stored by the source storage device between a start moment of the current backup period and an end moment of a previous backup period, and the target data block is different from all data blocks stored by the source storage device before the end moment of the previous backup period; obtaining the target data block according to the fingerprint information of the target data block; and sending the target data block to the backup storage device. Therefore, data backup efficiency can be improved, and consumption of computing resources and network resources can be reduced.