A Redundant Array Management Method and System Based on ZNS Solid State Drives
By leveraging the ZRWA feature on ZNS SSDs to achieve in-situ updates of parity blocks and lightweight metadata storage, the problem of in-situ updates of parity blocks on ZNS SSDs is solved, improving write performance and reducing latency, reducing garbage collection requirements, and ensuring data consistency and recovery efficiency.
CN119960673BActive Publication Date: 2026-05-26TSINGHUA UNIVERSITY
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TSINGHUA UNIVERSITY
- Filing Date
- 2024-12-27
- Publication Date
- 2026-05-26
AI Technical Summary
Technical Problem
Existing technologies make it difficult to update the checksum block in-situ on ZNS SSDs, resulting in low throughput, slow recovery speed, and frequent write amplification and garbage collection issues.
Method used
By using a redundant array management method based on ZNS solid-state drives, the ZRWA feature is utilized to achieve in-situ updates of the parity blocks, and lightweight metadata is stored in the spare area to ensure accurate location of logical write pointers after a failure, discard invalid data, and recalculate the parity data.
Benefits of technology
It improves write performance, reduces latency, reduces garbage collection requirements, ensures data consistency and recovery efficiency, and avoids additional storage space loss.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN119960673B_ABST
Abstract
This invention provides a redundant array management method and system based on ZNS solid-state drives, comprising: for each stripe, upon receiving a write request, sequentially writing the data to be written into the corresponding data blocks, and overwriting and updating the check data of the check block; positioning the logical actual write pointer to the tail position of the consecutively successfully written data blocks, generating metadata corresponding to each data block, and writing the metadata into a spare area; the logical actual write pointer is used to point to the next writable position within the logical area; in the event of a failure, repositioning the logical actual write pointer based on the metadata in the spare area; based on the repositioning of the logical actual write pointer after the failure, discarding the data blocks outside the logical actual write pointer, restoring the data blocks within the logical actual write pointer, and recalculating and generating check data to complete the redundant array management of the ZNS solid-state drive.
Need to check novelty before this filing date? Find Prior Art