Logical Slicing of Virtual Machine Backups for Instant Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in providing instant access to data stored in backup storage systems across multiple computing devices, leading to inefficiencies in data retrieval and management.
Innovation Solution
The proposed solution involves a method and system for managing data access by generating backups without crawling the file system, using metadata files for instant access, and employing resilient change tracking (RCT) to efficiently slice and restore large virtual machines (VMs).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional backup methods are used to backup large virtual machines, then complete data protection is achieved, but backup storage space consumption increases and data access time increases
Solution Approach 1:
The virtual machine disk is divided into multiple slices (e.g., slice 1, slice 2, slice 3) based on usage patterns and data importance. This segmentation allows the backup system to store only essential data in full while reducing the amount of data that needs to be backed up, thereby decreasing backup storage space consumption while maintaining data protection for critical information.
Solution Approach 2:
The patent extracts and separates frequently accessed data from less frequently accessed data. By identifying and extracting hot data (data with high access frequency) from cold data (data with low access frequency), the system can optimize backup strategies to prioritize storing hot data, reducing the overall backup volume while ensuring data protection for the most critical information.
2Reliability
If traditional backup methods are used, then complete data copying is achieved, but data retrieval time increases
Solution Approach 1:
By segmenting the virtual machine disk into multiple slices, the system enables parallel retrieval of data during restore operations. Multiple slices can be retrieved simultaneously from storage, significantly reducing the total data retrieval time compared to traditional sequential backup methods, while still maintaining complete data protection through the slice-based architecture.
Solution Approach 2:
The patent implements preliminary indexing and metadata creation during the backup process. Index structures are pre-built to map data locations, and metadata about each slice is stored to enable rapid data location and retrieval during restore operations, eliminating the need to scan through entire backup images sequentially and thus reducing data retrieval time.
3Productivity
If metadata is stored for each backup slice, then data access efficiency is improved, but metadata storage overhead increases
Solution Approach 1:
The patent applies local quality by creating metadata only for slices that contain frequently accessed data (hot data), while using simpler or no metadata for slices containing less frequently accessed data (cold data). This selective metadata approach reduces overall metadata storage overhead while maintaining high data access efficiency for the most critical data, optimizing the balance between productivity and storage resources.
Data Source
AI summary
A method for managing virtual machines obtaining a backup request for a full backup of a VM, in response to the backup request: performing a logical slicing of VM data associated with the virtual machine to obtain a set of logical slices, initiating generation of a backup of each logical slice in the set of logical slices to obtain a set of backup slices, storing the set of backup slices in a backup storage system, after storing the set of backup slices, obtaining a second backup request for a virtual synthetic backup of the VM, in response to the second backup request: obtaining a tracked changes file associated with the VM data after the full backup, identifying a backup slice associated with each of the tracked changes, and initiating storage of the virtual synthetic backup using the tracked changes file and the set of backup slices.