Replica Volume Data Paths with Segmented Writes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies and performance issues in disaster recovery (DR) sites due to the asynchronous replication of host data, where data is mostly written to replica volumes rather than read from them, leading to cache memory management challenges and suboptimal use of storage resources.
Innovation Solution
The solution involves partitioning large replication data transfers into small write requests, tagging these requests for cache management, deep compression, stream separation based on production site class of service, and flushing data to specific storage regions based on retention periods, thereby optimizing cache usage and storage efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If large replication data transfers are processed as single units, then data transfer efficiency is improved, but cache memory management becomes suboptimal and read penalties increase
Solution Approach 1:
The patent segments large replication data transfers into multiple smaller write requests. Each small write request is processed independently through the cache memory, allowing for optimized cache management. This segmentation enables the system to free up cache pages sooner and reduce read penalties while maintaining overall data transfer efficiency.
2Speed
If cache memory is used for storing replicated data, then write performance is improved, but cache memory becomes occupied longer and IO operations are delayed
Solution Approach 1:
The patent applies preliminary action by tagging small write requests with metadata that indicates they are replication operations. This allows the cache memory management system to proactively identify and evict these pages before they are needed for production IO operations, thereby reducing cache occupancy duration while maintaining write performance benefits.
Solution Approach 2:
The system uses feedback mechanisms where the cache memory manager receives information about replicated data writes and adjusts cache eviction policies accordingly. This feedback loop enables dynamic cache management that balances write performance with cache availability for production operations.
3Reliability
If replica volumes are used for disaster recovery, then data availability is improved, but storage resources are underutilized due to lack of reads
Solution Approach 1:
The patent applies local quality by differentiating the handling of replication data based on its destination. Small write requests to replica volumes are processed with specialized cache management techniques that recognize the data is intended for disaster recovery purposes. This allows the system to optimize cache usage for these specific operations while maintaining overall storage resource efficiency.
Data Source
AI summary
Techniques providing data path strategies for improving storage performance at DR sites. The techniques include receiving, in an asynchronous replication process, a large replication data transfer including data changes of a production volume since the last synchronization to a replica volume, partitioning the replication data into multiple small write requests, tagging each small write request as a write request to the replica volume, and performing early evicting, from cache memory, all cache pages used to cache host data specified in the small write requests; deep compression of contiguous host data specified in the small write requests; stream separation on the small write requests, each small write request being tagged as corresponding to a specific production site; and/or flushing host data having the same retention period to a specific region of physical storage space for the replica volume, each small write request being tagged with hint information indicating the retention period.