Asynchronous Replication via IO Splitter and Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data protection systems face challenges in maintaining continuous data protection without slowing down the production site, as they struggle to keep pace with high data transaction rates, leading to potential backlogs and system shutdowns due to the overhead of journaling processes.
Innovation Solution
The implementation of a method that intercepts IO operations in a replication environment, allowing them to be sent to a storage device before confirmation of receipt by a replication device, enabling asynchronous replication without the need for acknowledgement from the replication appliance, thereby reducing the overhead and allowing for dynamic adaptation to changing data transaction rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication with journaling is used to ensure data consistency, then data protection reliability is improved, but system performance deteriorates due to overhead and forced slowdowns at the production site
Solution Approach 1:
The patent segments the replication process into independent components: the production site continues processing data transactions without interruption, while a separate journaling process captures changes asynchronously. The splitter device divides the data stream and sends copies to both production storage and backup storage independently, allowing the production site to maintain high transaction rates without being constrained by backup operations.
Solution Approach 2:
The patent implements preliminary action by pre-positioning a splitter device in the data path that proactively captures and queues replication information before it becomes a bottleneck. The journaling process begins immediately when changes occur, using pre-allocated buffers and tracking structures to record modifications without waiting for backup storage confirmation, thereby preventing forced slowdowns at the production site.
2Productivity
If journaling overhead is reduced to maintain high data transaction rates, then productivity is improved, but data consistency and protection quality deteriorate
Solution Approach 1:
The patent introduces an intermediary journaling layer that acts as a buffer between the production site and backup storage. This journal captures all data changes with full detail including metadata, timestamps, and rollback information, ensuring complete data protection quality. The intermediary process handles the overhead of tracking and transmitting changes asynchronously, allowing the production site to operate at full speed without compromising protection quality.
3Reliability
If acknowledgement confirmation is required from backup storage before production site continues, then data consistency is improved, but downtime increases during replication operations
Solution Approach 1:
The patent implements dynamic replication where the system adapts its confirmation requirements based on operational context. During normal operations, replication proceeds asynchronously without requiring acknowledgements, minimizing downtime. The system dynamically adjusts by using sequence numbers and checksums to verify data integrity after the fact, rather than blocking operations for real-time confirmation, thus maintaining both consistency and high availability.
Data Source
AI summary
A method, system, and computer program product for use in a replication environment comprising intercepting an IO sent from a host to a storage device in a splitter, marking information about the IO in a tracking structure, sending the IO to the storage device, and sending the IO to a replication device; wherein the IO is sent to the storage device before confirming that the replication device has received the IO.