Incremental Image Backup Indexing Using File Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image backups, particularly those with high-density file systems, are resource-intensive and time-consuming due to the need for post-backup indexing, which involves parsing and categorizing the entire image content, making efficient retrieval of specific portions challenging.
Innovation Solution
A metadata-driven approach is employed during the backup process to generate indexing information, allowing for immediate post-backup indexing without rescanning the entire image, thereby reducing resource consumption and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional indexing methods are used after backup, then complete indexing of backup contents is achieved, but resource consumption and time required increase significantly
Solution Approach 1:
The patent applies preliminary action by generating indexing information during the backup process itself, before the backup is completed and stored. The data protection agent creates indexes of changed data blocks and their metadata while writing to backup storage, eliminating the need for post-backup rescanning and reducing resource consumption significantly.
Solution Approach 2:
The patent extracts only the necessary indexing information (changed data blocks and their metadata) during the backup process, rather than processing the entire backup image. This selective extraction of relevant data blocks for indexing reduces the computational burden while maintaining complete indexing capability.
2Measurement precision
If traditional indexing methods are used after backup, then complete indexing of backup contents is achieved, but time required for indexing increases significantly
Solution Approach 1:
The patent applies preliminary action by generating indexing information during the backup process itself, before the backup is completed and stored. The data protection agent creates indexes of changed data blocks and their metadata while writing to backup storage, eliminating the need for post-backup rescanning and reducing resource consumption significantly.
Solution Approach 2:
The indexing operation continues simultaneously with the backup writing process rather than occurring sequentially after backup completion. The data protection agent continuously creates and updates indexes as data blocks are written to backup storage, making the indexing action continuous and parallel with backup operations.
3Measurement precision
If full backup images are scanned for indexing, then all backup contents are indexed, but retrieval efficiency for specific portions decreases
Solution Approach 1:
The patent segments the backup image into individual changed data blocks, each with its own metadata and indexing information. This segmentation allows the indexing system to focus on discrete, manageable units rather than processing the entire backup image, improving retrieval efficiency for specific portions while maintaining completeness.
Solution Approach 2:
The patent extracts only the necessary indexing information (changed data blocks and their metadata) during the backup process, rather than processing the entire backup image. This selective extraction of relevant data blocks for indexing reduces the computational burden while maintaining complete indexing capability.
Data Source
AI summary
Techniques described herein relate to a method for performing backups of virtual machines on a host. The method includes obtaining an incremental image backup generation request associated with a virtual machine (VM) of the host; obtaining current file system metadata associated with the VM; obtaining previous backup metadata associated with the VM; generating updated file system metadata that includes status indicators associated with each file by comparing the current file system metadata with the previous backup metadata; obtaining a list of changed data blocks associated with the VM; writing changed data blocks associated with the VM to a backup storage; generating incremental backup metadata by associating portions of the updated file system metadata with each changed data block in the list of the changed data blocks; and writing the incremental backup metadata with the changed data blocks in the backup storage.