Incremental Backup for Online Availability Groups

Solution Overview

Problem

In on-line availability groups, frequent switches in active and passive database copies lead to inefficient creation of full backup files, wasting resources and exceeding backup windows due to lack of change records on passive nodes.

Innovation Solution

Agents inform each other of backup file creation, enabling change trackers to begin tracking changes from synchronizers, allowing incremental backup operations even on newly activated data object copies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If full backup files are created on every node after active-passive switches, then data protection is maintained, but resource wastage increases and backup windows are exceeded

Engineering Contradiction:

Improvedata protectionVSAvoidresource wastage

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The system performs preliminary actions by having the active node create backup files before switching to passive status, and by pre-establishing change tracking mechanisms on passive nodes. This ensures that when a node becomes passive, it already has the necessary backup information and change tracking capability, eliminating the need to create full backups after switching.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating backup files on the active node and distributing them to passive nodes, rather than creating independent full backups on each passive node. Passive nodes receive copies of the backup files and use change tracking to identify what needs to be backed up, significantly reducing redundant data creation and storage requirements.

Inventive Principle:

Principle #26Copying

2Reliability

If full backup files are created on every node after active-passive switches, then data protection is maintained, but backup completion time increases

Engineering Contradiction:

Improvedata protectionVSAvoidbackup window

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by having the active node create backup files before switching to passive status, and by pre-establishing change tracking mechanisms on passive nodes. This ensures that when a node becomes passive, it already has the necessary backup information and change tracking capability, eliminating the need to create full backups after switching.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by creating backup files on the active node and distributing them to passive nodes, rather than creating independent full backups on each passive node. Passive nodes receive copies of the backup files and use change tracking to identify what needs to be backed up, significantly reducing redundant data creation and storage requirements.

Inventive Principle:

Principle #26Copying

3Device complexity

If change tracking is not implemented on passive nodes, then system complexity is reduced, but incremental backup efficiency is lost

Engineering Contradiction:

Improvesystem complexityVSAvoidbackup efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The change tracking mechanism is designed to be universal and multi-functional. It can operate on both active and passive nodes, serving dual purposes: tracking changes for incremental backups and providing data for backup verification. This universal approach maximizes the utility of the change tracking capability across different node states without requiring separate mechanisms.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

Passive nodes perform self-service by autonomously tracking their own changes and identifying what needs to be backed up, rather than relying on the active node to manage all backup operations. This self-service capability reduces the overall system complexity by distributing the backup management workload and eliminating the need for complex centralized coordination.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10628444B1Incremental backup operations for on-line availability groups

Publication Date: 2020.04.21 EMC IP HLDG CO LLC

AI summary

A first agent informs a second agent that a backup file is created for an active copy of a data object in a first node in a cluster of nodes, in response to the first agent determining that the backup file is created for the active copy of the data object in the first node. The second agent instructs a change tracker associated with a passive copy of the data object in a second node in the cluster of nodes to begin tracking changes received from a synchronizer associated with the active copy of the data object in the first node, wherein tracked changes received from the synchronizer associated with the active copy of the data object in the first node enables a next backup operation performed on the data object to be performed as an incremental backup operation on the data object in the second node.