Global Application Data Maps for Accelerated Restoration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face high computational costs and inefficiencies in maintaining data redundancy and restoring data due to the lack of efficient indexing and metadata, leading to increased computational costs for data recovery.
Innovation Solution
A system utilizing a global application data map and deduplicated backup storage with virtual disk-level backups, enabling selective and efficient restoration of application data by associating backup data with applications without full indexing, reducing the amount of data needed for restoration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is stored in multiple locations to prevent data loss, then reliability is improved, but computational cost increases
Solution Approach 1:
The patent segments data into data blocks and creates a hierarchical backup structure where only changed blocks are backed up to remote locations. The backup system divides data into manageable units (data blocks) and uses a change block tracker to identify and backup only the modified portions, rather than copying entire datasets. This segmentation reduces the computational overhead of maintaining redundancy while preserving reliability through selective replication.
Solution Approach 2:
The patent extracts and tracks only the changed data blocks using a change block tracker, separating the backup process from the entire dataset. By extracting only the modified portions and storing metadata about data block locations and changes, the system maintains data redundancy without the computational cost of copying and managing complete data sets across multiple locations.
2Reliability
If complete data copies are maintained for restoration, then reliability is improved, but loss of time increases during restoration
Solution Approach 1:
The patent performs preliminary actions by maintaining a change block tracker and metadata that records data block locations, sizes, and change status before restoration is needed. This pre-computed information allows the restoration process to immediately identify and retrieve only the necessary changed blocks without scanning or analyzing complete data sets, significantly reducing restoration time while maintaining full restoration capability.
Solution Approach 2:
The patent extracts only the changed data blocks needed for restoration using the pre-maintained change block tracker metadata. Instead of retrieving complete data copies, the system identifies and extracts only the specific modified blocks that need to be restored, reducing restoration time while maintaining the ability to fully restore data to any previous state.
3Productivity
If full indexing is implemented for data access, then data access efficiency is improved, but device complexity increases
Solution Approach 1:
The patent segments the indexing approach by creating a lightweight metadata structure that tracks only data block locations and change status, rather than implementing comprehensive full-text or hierarchical indexing. This segmented metadata approach provides sufficient data access efficiency for backup and restoration operations without the complexity of extensive indexing structures.
Solution Approach 2:
The patent introduces a change block tracker as an intermediary metadata structure that mediates between the storage system and backup/restoration operations. This intermediary maintains minimal information (data block locations, sizes, and change status) that enables efficient data access and tracking without requiring complex indexing mechanisms, thus improving productivity while limiting device complexity.
Data Source
Figure 1
Figure 2.1
Figure 2.2
AI summary
A backup storage includes a persistent storage and a backup manager. The persistent storage stores backups of entities and a global application data map. The backup manager obtains a restoration request for restoration of an application hosted by an entity of the entities; makes a determination to accelerate the restoration based, at least in part, on the application; in response to the determination: identifies a virtual disk level backup of the entity that is associated with the application using the global application data map; obtains the virtual disk level backup from the backups of the entities; and restores the application using the virtual disk level backup.