Triangular Asynchronous Replication for Data Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data replication systems, such as synchronous and asynchronous RDF modes, face challenges with high latency and data consistency issues, especially over long distances, and require significant overhead in managing mirror volumes, leading to potential data loss during failures.
Innovation Solution
A system that toggles between storage configurations by disabling and enabling asynchronous data transfer between storage groups, using maps to associate synchronous and asynchronous data, and maintaining data integrity through chunk-based data transfer and mapping, allowing for efficient data recovery and synchronization across multiple storage destinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous RDF mode is used to ensure data consistency between primary and secondary storage devices, then data reliability is improved, but write operation latency increases due to waiting for acknowledgment
Solution Approach 1:
The patent segments the replication system into multiple independent storage groups (first, second, third storage groups) with different replication modes. This allows the system to divide data replication into synchronous and asynchronous paths, enabling low-latency writes to proceed asynchronously while critical data consistency is maintained through selective synchronous confirmation mechanisms.
Solution Approach 2:
The patent implements dynamic toggling between different storage configurations, allowing the system to switch between synchronous and asynchronous replication modes based on operational requirements. This dynamic adjustment enables the system to optimize between data consistency and write latency in real-time, resolving the contradiction by adapting the replication mode rather than being locked into a single approach.
2Speed
If semi-synchronous RDF mode is used to reduce write latency by acknowledging writes immediately, then write speed is improved, but data integrity is compromised because previous transfers must be acknowledged before new transfers begin
Solution Approach 1:
The patent segments data transfer into multiple independent storage groups that can operate in parallel. The first storage group handles synchronous transfers while the second storage group handles asynchronous transfers, allowing write operations to proceed without waiting for acknowledgment while maintaining data integrity through the segmented architecture.
Solution Approach 2:
The patent enables continuous data transfer by allowing the second storage group to perform asynchronous transfers without waiting for acknowledgment from the first storage group. This continuity of useful action eliminates the bottleneck in semi-synchronous mode where new transfers must wait for previous transfers to be acknowledged, thereby improving write speed while maintaining integrity through the segmented approach.
3Loss of time
If asynchronous RDF mode is used to minimize latency by copying data in the background, then write latency is reduced, but data consistency between primary and secondary storage devices cannot be guaranteed
Solution Approach 1:
The patent segments the replication system into multiple storage groups with different consistency guarantees. The first storage group provides strong consistency through synchronous transfers, while the second storage group provides high performance through asynchronous transfers. This segmentation allows the system to offer both low latency and data consistency depending on the specific replication path used.
Solution Approach 2:
The patent introduces a third storage group that acts as an intermediary between the first and second storage groups. This intermediary receives data from both sources and reconciles differences, ensuring data consistency is ultimately guaranteed even when asynchronous transfers are used, thereby resolving the contradiction between low latency and data consistency.
4Reliability
If multiple mirror volumes are maintained for disaster recovery, then data reliability is improved, but system complexity and overhead increase
Solution Approach 1:
The patent segments the disaster recovery system into multiple storage groups with different replication characteristics. This segmentation allows the system to maintain multiple copies of data (first storage group, second storage group, third storage group) without requiring a single complex management system, thereby reducing overhead while improving reliability through distributed redundancy.
Solution Approach 2:
The patent creates storage groups that can serve multiple functions: the first storage group can act as both a primary replication target and a backup source, while the second storage group can serve as both an asynchronous replication target and a failover source. This multi-functionality reduces the need for dedicated mirror volumes for each purpose, thereby reducing system complexity and overhead while maintaining disaster recovery capability.
Data Source
AI summary
Providing storage configurations for a data center includes providing synchronous data transfer from the data center to a local destination, establishing a first connection between the data center and a remote destination to provide chunks of data from the data center to a remote destination, establishing a second connection between the local destination and the remote destination to provide chunks of data from the local destination to the remote destination, and enabling at least one of the first and second connections to transfer chunks of data to the remote destination.