Forward-Incremental Backup Dataset Generation for Remote Replication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional backup systems face inefficiencies when replicating reverse-incremental backup datasets to remote secondary storage systems, resulting in redundant data transmission and storage due to the need to transmit full backups each time, which is inefficient and wasteful.

Innovation Solution

The method involves generating and maintaining forward-incremental backup datasets on secondary storage systems, which include remote full backups and forward incremental backups, reducing the need for redundant data transmission and storage by only transferring differences between data states.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If reverse-incremental backup datasets are replicated to remote secondary storage systems, then data reliability and fault-tolerance are improved, but data transmission efficiency deteriorates due to redundant full backup transmission

Engineering Contradiction:
Improvedata reliabilityVSAvoiddata transmission efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The backup dataset is segmented into a full backup dataset and multiple incremental backup datasets. The full backup contains all data at a specific point in time, while each incremental backup contains only the changes since the previous backup. This segmentation allows the system to transmit only the necessary portions (full backup initially, then smaller incremental updates) rather than repeatedly transmitting complete datasets, thereby improving transmission efficiency while maintaining reliability through the complete backup hierarchy.

Inventive Principle:
Principle #1Segmentation

2Reliability

If full backups are transmitted to remote storage each time data is backed up, then complete data protection is ensured, but storage resources are wasted due to redundant data

Engineering Contradiction:
Improvedata protectionVSAvoidstorage resources
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The backup system segments data protection into a full backup dataset and multiple incremental backup datasets. The full backup provides complete data protection at a specific point in time, while incremental backups provide efficient updates by storing only changes. This segmentation ensures complete data protection can be restored by combining the full backup with any necessary incremental backups, while minimizing storage resource consumption by avoiding redundant transmission and storage of unchanged data.

Inventive Principle:
Principle #1Segmentation

3Adaptability or versatility

If reverse-incremental backup is used, then restoration of prior states is enabled, but data transmission redundancy increases due to full backup repetition

Engineering Contradiction:
Improverestoration capabilityVSAvoiddata transmission redundancy
Core Design Contradiction:
Adaptability or versatilityVSLoss of substance

Solution Approach 1:

The patent inverts the traditional incremental backup approach by implementing forward incremental backups instead of reverse incremental backups. In forward incremental backup, each incremental backup is applied sequentially to reconstruct prior states, starting from an initial full backup. This inversion eliminates the need to repeatedly transmit full backups for prior state restoration, as each incremental backup contains only the changes since the previous backup, thereby reducing transmission redundancy while maintaining restoration capability.

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentUS10318386B1Systems and methods for maintaining remote backups of reverse-incremental backup datasets
Publication Date: 2019.06.11 COHESITY INC
  • US10318386B1 patent drawing
  • US10318386B1 patent drawing
  • US10318386B1 patent drawing

AI summary

A computer-implemented method for maintaining remote backups of reverse-incremental backup datasets may include (1) receiving a request to replicate a reverse-incremental backup dataset of a data source (e.g., a local full backup of the data source and/or at least one reverse incremental backup of the data source) from a primary storage system to a secondary storage system and (2) generating, in response to receiving the request, a forward-incremental backup dataset of the data source (e.g., a remote full backup of the data source and/or at least one forward incremental backup of the data source) on the secondary storage system based at least in part on the reverse-incremental backup dataset. Various other methods, systems, and computer-readable media are also disclosed.