Granular Cloud Directory Restore Across Storage Classes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional restore operations in cloud storage environments are limited to full volume, single file, or single directory restores, failing to support granular restoration of directories across multiple storage tiers, leading to inefficiencies and high operational costs.
Innovation Solution
A directory restore operation using base and instance metafiles for efficient management and checkpoint support, enabling selective restoration of directories across multiple storage tiers by tracking progress and restarting from checkpoints, and utilizing a rehydration mechanism to move objects from archival to standard storage tiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional restore operations are used (full volume, single file, or single directory), then the restore operation is simple to implement, but it cannot support granular restoration of directories across multiple storage tiers
Solution Approach 1:
The restore operation is segmented into multiple phases: traversal phase to identify objects across storage tiers, rehydration phase to move objects from archival to standard tier, and restore phase to actually restore data. This segmentation enables granular control while managing complexity through structured processing.
Solution Approach 2:
Before the actual restore operation, the system performs a traversal of the directory structure to identify all objects that need to be restored, including those in archival storage tiers. This preliminary action allows the system to plan the restore sequence and handle multi-tier complexity in advance.
2Quantity of substance
If objects are stored across multiple storage tiers (standard and archival), then storage cost is reduced, but restore operation fails when objects are in archival tier
Solution Approach 1:
The system introduces a rehydration mechanism as an intermediary step between accessing archival objects and the restore operation. Objects are first moved from archival to standard storage tier through rehydration, making them accessible to the restore operation. This resolves the reliability issue while maintaining cost efficiency.
Solution Approach 2:
The system performs a preliminary rehydration action before restore by moving objects from archival tier to standard tier. This ensures that all objects needed for restoration are in the accessible standard tier, guaranteeing restore operation success while maintaining the cost benefits of archival storage.
3Reliability
If full volume restore is performed, then all data is restored, but existing volume data is destroyed and operational cost increases
Solution Approach 1:
Instead of restoring the entire volume, the system extracts and restores only the specific directory and its contents that need recovery. This selective extraction approach maintains data restoration completeness for the target directory while avoiding the cost and data loss associated with full volume restores.
Solution Approach 2:
The system performs partial restore action by restoring only the necessary directory structure and files rather than the entire volume. This partial action reduces operational costs and prevents destruction of existing volume data while still achieving complete restoration of the target directory.
4Productivity
If single file restore is used, then only one file is restored, but it cannot restore subdirectories and nested structures
Solution Approach 1:
The system merges the restore operation with directory traversal functionality, allowing a single operation to handle both file restoration and subdirectory restoration. This combined approach maintains high restore speed while adding the capability to restore complete directory structures, resolving the contradiction between speed and versatility.
Data Source
AI summary
Techniques are provided for performing a storage operation targeting objects stored across multiple storage tiers of a cloud storage environment. A volume may be backed up as objects stored across the multiple storage tiers of the cloud storage environment, such as a standard storage tier directly accessible to the storage operation, an archival storage tier not directly accessible to the storage operation, etc. The storage operation may target the objects, such as where the storage operation is a directory restore operation to restore a directory of the volume. The storage operation can be successfully implemented such as to restore the directory even though objects of the storage operation are stored across the multiple storage tiers of the cloud storage environment.


