BCDR Data Replication via ZFS Snapshot Chains
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional business continuity and disaster recovery (BCDR) systems are complex and costly due to the difficulty in managing backup data and the significant computer resources they require, making them inefficient for minimizing the impact of computer system failures.
Innovation Solution
A method that replicates data blocks from a source computer to a target computer by capturing crash-consistent snapshots at different points in time, incrementally updating the target computer, and allowing it to take over the source computer's functions, using a Z File System (ZFS) chain for snapshot storage and management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional backup systems are implemented to protect against system failures, then data protection and system reliability are improved, but device complexity and resource consumption increase significantly
Solution Approach 1:
The backup system is segmented into three distinct components: a protected computing device that generates data, a recovery computing device that stores snapshots, and a target computing device that receives updates. This segmentation distributes complexity across multiple specialized devices rather than concentrating it in a single backup system.
Solution Approach 2:
Crash-consistent snapshots are captured in advance before potential system failures occur. These snapshots are stored on the recovery computing device, preparing backup data beforehand so that no time is lost during actual recovery operations.
Solution Approach 3:
Instead of traditional backup methods that copy entire file systems, the system creates block-level copies of data at specific points in time. These block copies are then incrementally updated and transferred to the target device, reducing the complexity of managing complete system backups.
2Reliability
If traditional backup systems are implemented to protect against system failures, then data protection is improved, but computer resource consumption increases significantly
Solution Approach 1:
Rather than backing up entire file systems, the system performs partial actions by capturing only the specific data blocks that have changed since the last snapshot. This incremental approach reduces the computational resources required for backup operations while maintaining data protection.
Solution Approach 2:
The system uses block-level copying instead of file-level copying, which is more efficient at the block level. Changed data blocks are identified and copied selectively, reducing the overall resource consumption compared to traditional backup methods that must process entire file systems.
3Productivity
If incremental updates are used to replicate data blocks, then productivity and efficiency are improved, but the complexity of managing snapshot chains increases
Solution Approach 1:
The snapshot management system is segmented into automated components that handle different aspects of snapshot chain management. The protected device captures snapshots, the recovery device stores and tracks them, and the target device applies updates, distributing the management complexity across the system architecture.
Solution Approach 2:
The system uses feedback mechanisms to track which data blocks have changed between snapshots. This feedback information guides the incremental update process, automatically determining what needs to be replicated without requiring manual intervention to manage the snapshot chains.
Data Source
AI summary
Disclosed systems, methods, and computer readable media relate to business continuity and disaster recovery (BCDR). According to some embodiments, disclosed systems, methods, and computer readable media can replicate a set of data blocks from a source computer to a target computer while the source computer is in operation and contents of the set of data blocks are changing. In some embodiments, disclosed systems, methods, and computer readable media can replicate a set of data blocks from a recovery computer to a target computer while the recovery computer is in operation and contents of the set of data blocks are changing, wherein the recovery computer is built from a protected computer.