Collision Handling in Asynchronous Replication via Sequence Tagging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Synchronous replication introduces latency due to the need for acknowledgement from the destination array, while asynchronous replication faces issues like scrambled data, overwrite risks, and lost transactions due to link or node failures, and write ordering challenges across distributed systems.
Innovation Solution
The solution involves using a distributed set management system where transactions are tagged with a cluster-wide correlator and grouped into sets based on common properties, allowing for efficient processing and sequencing across distributed platforms, with mechanisms for collision handling and node failure recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is used to ensure data integrity, then reliability is improved, but speed deteriorates due to acknowledgement latency
Solution Approach 1:
The patent segments the replication process into independent transaction units that can be processed asynchronously. Each transaction is assigned a sequence number and replicated independently without blocking, allowing the system to maintain data integrity through sequencing while improving replication speed by eliminating acknowledgement latency.
Solution Approach 2:
The patent implements preliminary actions by pre-assigning sequence numbers to transactions before replication and maintaining a write ordering mechanism at the destination. This allows the destination array to process transactions in the correct order without requiring real-time acknowledgements, thus preserving reliability while enhancing speed.
2Speed
If asynchronous replication is used to improve speed, then replication speed is improved, but reliability deteriorates due to scrambled data and write ordering issues
Solution Approach 1:
The patent implements a feedback mechanism through sequence numbers assigned to each transaction. The destination array uses these sequence numbers to verify and maintain the correct write order, providing feedback on transaction processing status without requiring blocking acknowledgements. This ensures reliability while maintaining asynchronous speed.
Solution Approach 2:
The patent changes the parameter of transaction identification from simple acknowledgements to sequence-numbered transactions. This parameter change allows the system to track and enforce write ordering asynchronously, resolving the reliability issue while preserving the speed benefit of asynchronous replication.
3Reliability
If a distributed set management system is implemented to handle collisions, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where each transaction carries its own sequence number and collision information. The destination array autonomously processes collisions by comparing sequence numbers without requiring complex external coordination. This maintains reliability through proper collision handling while minimizing system complexity by distributing the management logic to individual transactions.
Data Source
AI summary
Methods and systems for collision handling during an asynchronous replication are provided. A system includes a cache memory system comprising a number of cache memory pages. A collision detector detects when a host is attempting to overwrite a cache memory page that has not been completely replicated. A revision page tagger copies the cache memory page to a free page and tags the copied page as protected.