Persistent Cache Mirrored RAID Write Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
RAID configurations that utilize data mirroring suffer from increased latency due to the need to write the same data on multiple devices, with the global request not being reported as completed until the slowest member finishes its write request, which affects performance and redundancy.
Innovation Solution
Implementing a persistent cache with low write access latency, caching all writes to one member persistent storage drive in a striped manner and synchronizing data across drives when the array is idle, while maintaining data redundancy through mirroring.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written to multiple drives simultaneously for mirroring, then data redundancy is improved, but write I/O latency increases
Solution Approach 1:
The patent applies preliminary action by writing data to the persistent cache before writing to the mirrored drives. The controller caches the write data in the persistent cache first, then asynchronously writes to the mirrored drives in the background. This allows the system to acknowledge the write operation immediately after caching, rather than waiting for all mirrored drives to complete their writes, thus reducing write I/O latency while maintaining data redundancy through the cache.
2Loss of time
If data is cached in persistent cache before writing to drives, then write I/O latency is reduced, but data redundancy mechanism changes
Solution Approach 1:
The persistent cache acts as an intermediary between the host system and the mirrored drives. Instead of directly writing to multiple drives simultaneously, the controller first writes data to the persistent cache, which then serves as a buffer and coordination point. The cache enables asynchronous write operations to mirrored drives while providing a single point of control for managing redundancy, thus reducing latency while preserving the mirroring redundancy mechanism.
3Stability of the object's composition
If write operations wait for slowest drive completion, then data consistency is improved, but overall system performance deteriorates
Solution Approach 1:
The system performs the critical data consistency action by first writing to the persistent cache, which has faster write speeds. This preliminary write ensures data is safely stored before initiating background writes to slower mirrored drives. The controller can immediately acknowledge the write operation to the host system after the cache write completes, rather than waiting for the slowest drive, thus maintaining data consistency through the cache while improving overall system performance.
Data Source
AI summary
An embodiment of a semiconductor apparatus for use with redundant storage may include technology to cache all data for a write request for at least two member persistent storage drives in a persistent cache with a write access latency at least as low as a lowest write access latency of the at least two member persistent storage drives, write the data for the write request to one member persistent storage drive of the at least two member persistent storage drives, and indicate that the write request is complete after the data for the write request is redundantly stored in the persistent cache and the one member persistent storage drive. Other embodiments are disclosed and claimed.