Parallel Virtual Disk Backup via Split BAT Plexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional backup methodologies for virtual disk files are inefficient, leading to increased backup windows and underutilization of network resources, failing to meet Recovery Time Objectives (RTOs) and Recovery Point Objectives (RPOs, and causing contention with business applications.
Innovation Solution
The method involves splitting the block allocation table (BAT) of a virtual disk file into multiple split BATs, assigning data blocks to plexes, and streaming these plexes in parallel to a backup apparatus, utilizing multiple network streams to improve resource utilization and speed up the backup process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional single-stream backup methodology is used, then network resource utilization is low, but backup time is excessive and does not meet RTO/RPO objectives
Solution Approach 1:
The patent segments the backup process by splitting the block allocation table (BAT) into multiple split BATs and assigning data blocks to different plexes, enabling parallel streaming of backup data across multiple network streams simultaneously, thereby increasing backup speed and reducing backup window
Solution Approach 2:
The patent transitions from a single-dimensional sequential backup stream to a multi-dimensional parallel streaming architecture, where multiple plexes stream data simultaneously across different network channels, effectively adding temporal and spatial dimensions to the backup process
2Loss of time
If snapshot capability is used to enable backups during production, then downtime is reduced, but compute resource contention with business applications occurs
Solution Approach 1:
The patent extracts the backup operation from the production workload by using snapshots to capture data at a specific point in time, then performing the actual backup of snapshot data separately, allowing backups to proceed without interfering with business application compute resources
Solution Approach 2:
The patent performs preliminary action by creating snapshots before the backup process, capturing the desired data state in advance, which allows the subsequent backup operation to work with static snapshot data rather than live production data, eliminating resource contention
3Reliability
If large data sets (TBs or more) are backed up using conventional technologies, then data integrity is maintained, but backup window increases significantly
Solution Approach 1:
The patent segments large data sets into smaller manageable units organized through split BATs and plexes, allowing parallel processing of multiple data segments simultaneously, which maintains data integrity through structured organization while dramatically reducing the overall backup window for terabyte-scale data
Data Source
AI summary
One embodiment is related to a method for backing up a virtual disk file using parallel save streams, comprising: retrieving a block allocation table (BAT) of the virtual disk file; splitting the BAT into a plurality of split BATs; assigning data blocks of the virtual disk file to plexes based on the plurality of split BATs, wherein each plex corresponds to one of the plurality of split BATs; and streaming the plexes to a backup apparatus in parallel, wherein each plex occupies one stream.