DDDR Backup Tracking Segmentation for Ungraceful Shutdown Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing backup systems face challenges in efficiently recovering tracking information for incremental backups after an ungraceful shutdown, such as power loss, leading to resource-intensive comparisons of all data blocks to identify changed blocks.
Innovation Solution
The Dynamically Determined Difference Regions Apparatuses, Methods, and Systems (DDDR) implement a DDDR component that tracks changed data blocks by identifying ranges containing changed data blocks, using hash values or checksums, and updates Changed Block Tracking (CBT) system information, thereby reducing the need for comprehensive comparisons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comprehensive comparison of all data blocks is performed to identify changed blocks after ungraceful shutdown, then tracking information can be recovered, but resource consumption increases significantly
Solution Approach 1:
The patent segments the data storage space into multiple regions and uses bitmaps to track changed blocks in each region separately. After an ungraceful shutdown, the system only needs to compare and recover tracking information for specific regions where changes are likely to have occurred, rather than performing a comprehensive comparison of all data blocks. This segmentation approach maintains reliable tracking information recovery while significantly reducing the computational resources and time required.
2Reliability
If comprehensive comparison of all data blocks is performed to identify changed blocks, then accurate incremental backups can be achieved, but time consumption increases
Solution Approach 1:
The patent implements preliminary action by maintaining bitmaps that continuously track changed blocks during normal operation before any shutdown occurs. These bitmaps are periodically flushed to persistent storage. After an ungraceful shutdown, the system can directly recover and use the pre-computed bitmap information to identify changed blocks, eliminating the need for time-consuming comprehensive comparisons and ensuring accurate incremental backups are achieved efficiently.
3Loss of information
If all data blocks are compared to identify changed blocks, then complete tracking information is obtained, but device complexity increases
Solution Approach 1:
The patent introduces bitmaps as intermediary data structures that mediate between the data blocks and the tracking system. Each bitmap serves as a compact representation of change status for a specific region, allowing the system to obtain complete tracking information without directly comparing all individual data blocks. This intermediary approach maintains information completeness while reducing system complexity by providing a simplified mechanism for tracking and recovering change information.
Data Source
AI summary
The Dynamically Determined Difference Regions Apparatuses, Methods and Systems (“DDDR”) transforms backup configuration request, backup data read response inputs via DDDR components into backup configuration response, changed blocks update request outputs. A write command for a data block to write to a data volume is detected. A data volume address of the data block is determined. A superset range associated with the data volume address is determined. Upon determining that the written-to flag of the superset range is set: extend a previously established written-to subset range within the superset range to include the data volume address when it is determined that the data volume address is not within the previously established written-to subset range. Upon determining that the written-to flag is not set: set the written-to flag and establish a new written-to subset range within the superset range that includes the data volume address. Execute the write command.