Virtual Machine Backup Consistency via File-Level Segmentation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing backup and restore applications for virtual machines restore data sets on an individual basis, leading to inconsistent recovery of multiple virtual machines in case of file failure, resulting in loss of changes for some virtual machines due to differing backup policies.

Innovation Solution

A backup and restore application identifies a file storing multiple virtual machines, creates backups using change block tracking, and restores all virtual machines based on the most recent backup of the data store, even if some virtual machines were not registered or had outdated backup policies, by responding to user input and ensuring sufficient storage capacity for restoration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If backup policies are applied individually to each virtual machine, then each virtual machine can have customized backup schedules, but recovery consistency across multiple virtual machines is lost when a file failure occurs

Engineering Contradiction:

Improvebackup policy customizationVSAvoidrecovery consistency

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the backup process into two levels: file-level backup (backup entire data store containing multiple VM files) and VM-level restore (selective restoration of individual virtual machines). This segmentation allows the backup policy to be applied uniformly at the file level while enabling customized restoration at the VM level, resolving the contradiction between backup policy flexibility and recovery consistency.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary restoration process that sits between the backup storage and the virtual machines. When a file failure occurs, the system first restores the affected file to a temporary location, then selectively restores individual VMs from that temporary location based on their specific backup policies. This intermediary step ensures both recovery consistency (all VMs can be restored from the same backup point) and policy customization (each VM follows its own restore schedule).

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If individual virtual machine restoration is performed, then backup policies can be selectively applied, but data loss occurs for virtual machines with outdated or no backup policies

Engineering Contradiction:

Improveselective restorationVSAvoiddata loss

Core Design Contradiction:

Ease of operationVSLoss of information

Solution Approach 1:

The patent performs preliminary restoration of the failed file to a temporary location before attempting VM-level restoration. This preliminary action ensures that the base data is recovered first, preventing information loss. Then, VMs are selectively restored from this temporary location, ensuring that even VMs with outdated or no backup policies can be recovered from the most recent available backup point.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system creates a cushioning layer by restoring the failed file to a temporary location before VM-level restoration. This temporary restoration acts as a safety buffer, ensuring that data is preserved in an intermediate state. This cushioning mechanism prevents information loss by providing a fallback restoration point for all VMs, regardless of their individual backup policy status.

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

3Reliability

If all virtual machines are restored from the most recent backup, then data consistency is maintained, but restoration time increases for machines that don't need full restoration

Engineering Contradiction:

Improvedata consistencyVSAvoidrestoration time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts only the necessary VMs for restoration from the set of all VMs. After the preliminary file restoration to temporary location, the system identifies which specific VMs need restoration based on their backup policies and failure impact, then restores only those extracted VMs. This extraction approach maintains data consistency for affected VMs while minimizing unnecessary restoration operations and reducing overall restoration time.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs partial restoration by selectively restoring only the VMs that are impacted by the file failure or require updates, rather than restoring all VMs. This partial action approach maintains consistency for the necessary VMs while avoiding excessive restoration operations on VMs that don't need restoration, thereby reducing time loss.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10853201B1Backing up files storing virtual machines

Publication Date: 2020.12.01 EMC IP HLDG CO LLC

AI summary

Backing up files storing virtual machines is described. A system identifies a file storing multiple virtual machines. The system creates a backup of the file. The system identifies a failure of the file. The system restores the multiple virtual machines based on the backup of the file.