Asynchronous Storage Synchronization via Activity-Based Extent Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data protection systems face inefficiencies in synchronizing data between primary and secondary storage volumes, particularly in asynchronous replication systems, leading to unnecessary data transfer and prolonged synchronization times, which can increase the recovery point objective (RPO) and hinder local reads during synchronization.
Innovation Solution
The approach synchronizes storage volumes by prioritizing extents based on activity levels, delaying synchronization of write-active areas and focusing on less active regions, thereby reducing data transfer and synchronization time, and allowing efficient local reads from the replica site.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data protection systems synchronize all data blocks between primary and secondary storage volumes, then data consistency is maintained, but synchronization time increases and recovery point objective (RPO) worsens
Solution Approach 1:
The storage volume is divided into multiple data blocks, allowing selective synchronization of only those blocks that have changed since the last synchronization event. This segmentation enables the system to maintain data consistency for critical blocks while avoiding unnecessary transmission of unchanged blocks, thereby reducing synchronization time and improving RPO.
Solution Approach 2:
Instead of synchronizing all data blocks uniformly, the system performs partial synchronization by identifying and transmitting only the subset of blocks that require updating. This partial action approach maintains essential data consistency while significantly reducing the time and bandwidth required for synchronization operations.
2Reliability
If conventional systems transmit all changed data blocks over the network, then data integrity is preserved, but the amount of data transmitted increases
Solution Approach 1:
The system extracts and identifies only those data blocks that have changed since the last synchronization, separating them from the unchanged majority. By transmitting only these extracted changed blocks rather than all blocks, the system preserves data integrity for updated portions while dramatically reducing the total volume of data transmitted over the network.
Solution Approach 2:
The system changes the parameter of data transmission from transmitting all blocks to transmitting only changed blocks. This parameter change is achieved through tracking mechanisms that identify modified blocks, allowing the system to maintain data integrity while optimizing network bandwidth utilization by reducing transmission volume.
3Reliability
If synchronous replication is used to maintain real-time data consistency, then RPO is minimized, but system complexity and performance overhead increase
Solution Approach 1:
The system implements dynamic replication where the synchronization behavior adapts based on the state of individual data blocks. Rather than applying a static synchronous or asynchronous model to all data, the system dynamically determines which blocks require synchronization and transmits only those, reducing overall system complexity while maintaining effective RPO through selective real-time updates of changed blocks.
Data Source
AI summary
In one aspect, a method includes synchronizing a second storage volume to a first storage volume by synchronizing extents according to priority based on extent activity. In another aspect, an apparatus includes electronic hardware circuitry configured to synchronize a second storage volume to a first storage volume by synchronizing extents according to priority based on extent activity. In a further aspect, an article includes a non-transitory computer-readable medium that stores computer-executable instructions. The instructions cause a machine to synchronize a second storage volume to a first storage volume by synchronizing extents according to priority based on extent activity.