Timestamp Logic for Synchronous Replication Data Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In virtualized storage clusters, maintaining data consistency across mirrored nodes is challenging due to failures such as network link failures, hardware failures, and power outages, especially when using write-back caches, leading to inconsistencies and requiring resynchronization, which can introduce overhead and latency.
Innovation Solution
Utilizing timestamps as a watermark to determine the re-synchronization point in both primary and secondary nodes, ensuring that data consistency is maintained by synchronizing based on the most recent access time and incrementing a time segment number, and prioritizing re-synchronization of territories with lesser time segment numbers to minimize data re-synced.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is used to maintain data consistency across mirrored nodes, then data integrity is improved, but system complexity and overhead increase due to the need for resynchronization after failures
Solution Approach 1:
The patent replaces complex mechanical resynchronization mechanisms with a timestamp-based logical system. Instead of tracking and resynchronizing individual data blocks after failures, the system uses timestamps to identify the most recent consistent state, substituting a simple time-based marker for complex data comparison and reconciliation processes.
Solution Approach 2:
The patent changes the parameter used for tracking data state from detailed content tracking to a simplified timestamp parameter. By monitoring only the timestamp of the last successful replication rather than the actual data contents, the system reduces the complexity of detecting and resolving inconsistencies after failures.
2Productivity
If write-back cache is used to improve write performance, then I/O speed is improved, but data consistency is worsened due to potential data loss during power failures
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously monitors timestamps from both primary and secondary nodes. After a power failure or interruption, the system uses this timestamp feedback to automatically detect inconsistencies and trigger appropriate resynchronization actions, allowing write-back cache to be used safely while maintaining consistency.
Solution Approach 2:
The patent performs preliminary timestamp recording at the point of last successful replication. This preliminary action creates a reference point that can be used after failures to determine what data needs to be resynchronized, preventing data loss issues associated with write-back cache by having the recovery information prepared in advance.
3Reliability
If full resynchronization is performed after node failures, then data consistency is improved, but time and performance are worsened due to redundant data transfer
Solution Approach 1:
The patent extracts only the essential information needed for resynchronization - the timestamp of the last successful replication - rather than transferring or comparing all data contents. This extraction approach allows the system to quickly identify what needs to be resynchronized without the overhead of examining or transferring redundant data that is already consistent.
Solution Approach 2:
The patent implements partial resynchronization by using timestamps to identify only the specific data portions that need to be synchronized, rather than performing excessive full resynchronization of all data. This partial action approach transfers only the necessary data after failures, reducing time loss while ensuring consistency.
Data Source
AI summary
Systems and methods for maintaining data consistency across mirrored storage nodes. Data consistency may be maintained in networked storage environments using a timestamp of an I/O in the primary and secondary servers (also referred to as nodes) in synchronous replication to re-synchronize the data efficiently, without much performance drop and while reducing the memory footprint. According to an aspect presented herein, data consistency is maintained in a mirror node by timestamping the I/O for a volume's logical territory in both the primary and secondary nodes. The timestamp acts as a watermark. Thus, whenever a disaster or failure occurs, the watermark point determines the re-synchronization point.