Replication Based Security for Data Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data protection systems face challenges in minimizing downtime during recovery and keeping pace with high data transaction rates, leading to potential production site shutdowns due to the overhead of journaling, which can result in data loss and prolonged recovery times.
Innovation Solution
A computer-implemented method intercepts production I/O operations at a splitter, replicates write I/O and metadata of read I/O, allowing for continuous data protection by maintaining a journal of changes, enabling rollback to any point in time and detecting unauthorized access or data theft.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If journaling is used for continuous data protection, then recovery point in time is improved, but data transaction throughput deteriorates
Solution Approach 1:
The patent segments I/O operations into two distinct paths: a fast path for read operations that bypasses journaling overhead, and a slow path for write operations that goes through journaling for protection. This segmentation allows the system to maintain high throughput for reads while ensuring protection for writes, resolving the contradiction between recovery capability and transaction throughput.
Solution Approach 2:
The patent applies different handling qualities to different I/O operations locally. Read operations receive fast, direct handling without journaling overhead, while write operations receive protected handling with journaling. This local differentiation allows the system to optimize for both performance and protection at the operation level, eliminating the need to slow down the entire system for protection.
2Reliability
If synchronous replication is used for continuous protection, then data protection reliability is improved, but production site performance deteriorates
Solution Approach 1:
The patent implements dynamic replication mode switching based on operational conditions. The system can transition between synchronous and asynchronous replication modes, allowing it to maintain reliability during critical moments while optimizing performance during normal operation. This dynamic adaptation resolves the contradiction by making replication flexibility rather than fixed.
Solution Approach 2:
The patent employs periodic or event-driven replication updates rather than continuous synchronous replication. Changes are replicated periodically or when specific events occur, reducing the performance impact on the production site while maintaining adequate protection reliability. This periodic approach eliminates the need for continuous synchronization that would throttle production performance.
3Reliability
If read I/O is replicated to maintain continuous protection, then data integrity is improved, but replication overhead increases
Solution Approach 1:
The patent extracts read operations from the journaling path, allowing them to proceed independently without contributing to replication overhead. By taking reads out of the protected path, the system maintains data integrity for writes while avoiding the unnecessary replication overhead that would result from replicating every read operation. This extraction resolves the contradiction by selectively protecting only what is necessary.
Data Source
AI summary
A computer implemented method, system, and computer comprising intercepting an production IO at a splitter, determining if the production IO is a write IO, based on a positive determination of a write IO; replicating a copy of the write IO, based on a negative determination, determining if the production IO is a read IO and based on a positive determination of a read IO; replicating to metadata of the read IO.