Continuous Data Verification for Synchronous Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In information processing systems, synchronous replication often leads to data inconsistencies between source and target storage systems due to malfunctions like link loss or code issues, which are typically detected too late to correct, causing synchronization loss and impacting system performance.
Innovation Solution
Implementing continuous data verification during synchronous replication by using content-based signatures and secure hashing algorithms to detect and correct inconsistencies in real-time without affecting storage system performance, utilizing a clustered content addressable storage system with distributed storage controllers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is implemented without continuous verification, then storage system performance is maintained, but data inconsistency between source and target systems occurs due to malfunctions like link loss or storage module restart
Solution Approach 1:
The system performs preliminary actions by continuously generating and comparing content-based signatures of data pages before inconsistencies can propagate. The source storage system continuously creates signatures of source data pages and transmits them to the target storage system for verification, detecting inconsistencies at an early stage before they impact overall system reliability
Solution Approach 2:
The system implements feedback mechanisms where the target storage system continuously verifies received signatures against locally computed signatures and reports verification results back to the source. This closed-loop feedback enables real-time detection and correction of synchronization issues without requiring system-wide performance degradation
2Reliability
If continuous data verification is implemented, then data inconsistency is detected and corrected in real-time, but storage system performance may be adversely impacted
Solution Approach 1:
The system extracts only the essential verification information (content-based signatures) from the full data set and transmits only these compact representations for verification. By separating the verification function from the data itself and transmitting only signature data rather than full data pages, the system minimizes network overhead and processing requirements while maintaining continuous verification capability
Solution Approach 2:
The system changes the verification parameter from comparing entire data pages to comparing compact content-based signatures. This parameter transformation reduces the computational complexity and data transmission requirements from O(n) to O(1) per page, enabling continuous verification without adversely impacting storage system performance
3Device complexity
If data verification is performed only after disaster occurs, then system complexity is minimized, but data inconsistency cannot be corrected in time
Solution Approach 1:
The system performs preliminary verification actions continuously in the background before disasters occur. By continuously generating, transmitting, and verifying content-based signatures during normal operation, the system detects and corrects inconsistencies at an early stage rather than waiting for disaster scenarios, maintaining simplicity while improving reliability
Data Source
AI summary
An apparatus includes a processing device comprising a processor coupled to a memory, with the processing device being configured, in conjunction with synchronous replication of at least one logical storage volume between first and second storage systems, to acquire an address lock for a set of pages of the logical storage volume starting from a particular page of the logical storage volume, to determine content-based signatures for respective pages of the set of pages, to compute an additional signature as a function of the content-based signatures, and to send the additional signature and a pointer to the particular page from the first storage system to the second storage system, so as to permit the second storage system to verify consistency of the set of pages in the second storage system relative to the set of pages in the first storage system. Such operations are repeated for other sets of pages.