Parallel Backup Cloning Across Multiple Hosts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data backup cloning processes are slow and resource-intensive due to their sequential nature, particularly when dealing with large datasets, and are hindered by the complexity of formats like VHD and VHDx, which are space-efficient but difficult to access traditionally.
Innovation Solution
Implementing parallel cloning techniques across multiple hosts that partition and clone portions of a primary backup in parallel, combined with a disk file conversion mechanism to convert sparse disk files into native disk files for efficient access, allowing simultaneous cloning operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cloning is performed sequentially to create multiple backup copies, then data redundancy is improved, but cloning time and resource consumption increase significantly
Solution Approach 1:
The patent divides the primary backup into multiple data portions and assigns each portion to a different host for parallel cloning operations. This segmentation allows simultaneous processing of multiple clone operations, reducing overall cloning time while maintaining data redundancy through multiple backup copies.
Solution Approach 2:
The patent transitions from sequential single-host cloning to parallel multi-host cloning by adding the dimension of concurrent processing. Multiple hosts operate simultaneously on different data portions, transforming the cloning process from a linear sequence to a parallel operation that reduces time consumption.
2Quantity of substance
If VHD or VHDx formats are used for space-efficient backup storage, then storage efficiency is improved, but access complexity and difficulty increase
Solution Approach 1:
The patent introduces a conversion mechanism that transforms VHD/VHDx sparse disk files into native disk formats. This intermediary conversion process resolves the access complexity by converting the proprietary format into a universally accessible format, while the underlying sparse structure maintains storage efficiency.
Solution Approach 2:
The patent changes the format parameter of the backup data from VHD/VHDx to native disk format through conversion. This parameter change enables traditional access techniques to be applied while preserving the space-efficient characteristics of the original sparse disk file structure.
3Reliability
If twinning is used to create two backup copies simultaneously, then data redundancy is improved, but primary backup time increases to match the slower remote copy speed
Solution Approach 1:
The patent segments the backup data and distributes portions to multiple hosts that perform cloning in parallel. This allows the primary backup to complete faster without waiting for all clone operations to finish, as each host works independently on its assigned data portion.
Solution Approach 2:
The patent implements partial cloning where multiple hosts clone different portions of the backup data simultaneously. This partial parallel action allows the primary backup to proceed at full speed while clone operations occur concurrently, rather than waiting for complete duplication.
Data Source
AI summary
A request is received to clone a primary backup copy. An identification is made of a disk image file including the primary backup copy. The disk image file is scanned to split an amount of data to be cloned among a set of hosts. The hosts are provisioned. The hosts are assigned a portion of the amount of data to be cloned. A host clones its respective portion in parallel with another host to create clone sets on a clone target. A clone set includes a copy of at most a portion of the primary backup copy.