Virtual Machine Restore via Predictive Data Pre-staging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data management systems face challenges in efficiently restoring virtual machines from backup to production environments while minimizing downtime and managing the costs associated with explosive data growth.
Innovation Solution
The system employs predictive analysis to pre-stage backed-up data blocks in a read cache, allowing for rapid booting and execution of virtual machines, with the option to relocate data to primary storage or maintain access in secondary storage for quick recovery, using an enhanced data agent and media agent within a storage management system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If virtual machines are backed up to secondary storage for cost-effective data protection, then storage costs are reduced, but restore time increases
Solution Approach 1:
The system performs preliminary actions by pre-staging data blocks from secondary storage to a read cache before the actual restore operation is needed. This advance preparation reduces the time required when restore is initiated, while maintaining cost-effective secondary storage as the primary backup location.
Solution Approach 2:
A read cache component is introduced as an intermediary between secondary storage and the virtual machine execution environment. This intermediary layer holds pre-staged data blocks, enabling fast access during restore operations without requiring continuous access to the slower secondary storage medium.
2Speed
If all backed up data blocks are pre-staged to read cache for instant VM restore, then restore speed increases, but storage resource consumption increases
Solution Approach 1:
Instead of uniformly pre-staging all data blocks, the system applies local quality by selectively pre-staging only those data blocks that are predicted to be needed for the restore operation. This selective approach optimizes restore speed for critical data while conserving read cache resources.
Solution Approach 2:
The system performs partial pre-staging of data blocks rather than complete pre-staging of all backup data. This partial action is sufficient to achieve instant VM restore functionality while avoiding the excessive consumption of read cache resources that would result from pre-staging all data blocks.
3Speed
If virtual machine files are relocated from backup to primary storage, then access performance improves, but data management complexity increases
Solution Approach 1:
The system implements dynamic data management where virtual machine files can be relocated from backup storage to primary storage based on operational needs. This dynamic approach allows the system to adapt data locations for optimal performance while maintaining the flexibility to revert to or maintain backup locations, managed through automated processes that reduce manual complexity.
Data Source
AI summary
Systems and methods enable a virtual machine, including applications executing thereon, to quickly start executing and servicing users based on pre-staged data blocks supplied from a backup copy in secondary storage. Substantially concurrently with the ongoing execution of the virtual machine, a virtual-machine-file-relocation operation moves data blocks from backup to a primary storage destination that becomes the virtual machine's primary data store after relocation completes. An enhanced data agent, operating in conjunction with an enhanced media agent in a storage management system, coordinates restoring the virtual machine and launching the relocation operation. The enhanced media agent pre-stages certain backed up data blocks based on predictive analysis pertaining to the virtual machine's operational profile. The enhanced media agent also pre-stages backed up data blocks for the relocation operation, based on the operation's relocation scheme. The media agent tracks read requests and maintains the contents of an associated read cache.