Parity Disk Data Write in Degraded RAID Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
RAID systems in degraded states experience inefficient I/O operations due to the need for parity information calculation and data reconstruction during writes and reads, leading to cumbersome and resource-intensive processes.
Innovation Solution
Directly writing data to a parity disk in the degraded state instead of calculating parity information, and setting corresponding degraded storage position information to facilitate direct data retrieval from the parity disk, reducing the need for complex calculations and reconstructions during reads and writes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If I/O operations are performed in degraded state using traditional RAID methods, then fault tolerance is maintained, but I/O efficiency deteriorates due to parity calculation and data reconstruction requirements
Solution Approach 1:
The patent extracts the parity disk from its traditional role of only storing parity information and enables it to also store data directly. When a data disk fails, the system writes data directly to the parity disk instead of performing parity calculations, thereby eliminating the computational overhead while maintaining fault tolerance through the degraded state management mechanism.
Solution Approach 2:
The patent inverts the traditional RAID write operation in degraded state. Instead of calculating parity and writing to the parity disk, or reading from multiple disks and reconstructing data, the system directly writes data to the parity disk location, reversing the conventional approach to achieve simpler and more efficient I/O operations.
2Reliability
If parity information calculation is performed during write operations in degraded state, then data integrity is maintained, but CPU resource consumption increases
Solution Approach 1:
The patent removes the parity calculation step from the write operation process in degraded state. By allowing direct data writing to the parity disk without computing parity information, the system eliminates CPU-intensive mathematical operations while maintaining data integrity through the degraded state management and metadata tracking mechanisms.
3Reliability
If data reconstruction is performed during read operations in degraded state, then correct data retrieval is ensured, but access time increases
Solution Approach 1:
The patent performs preliminary actions by maintaining metadata that tracks which data is stored on which disks, including the parity disk. When reading data in degraded state, the system consults this pre-established metadata to directly locate data on the parity disk without needing to perform time-consuming reconstruction calculations, thereby reducing access time while ensuring data accuracy.
4Reliability
If traditional RAID degraded state operations are used, then fault tolerance is maintained, but system complexity increases due to multiple read/write paths
Solution Approach 1:
The patent extracts and eliminates the complex parity calculation and data reconstruction logic from the degraded state operation paths. By enabling direct write to parity disk and using metadata-based direct read, the system reduces the number of operational paths and simplifies the control logic while maintaining fault tolerance through the degraded state management framework.
Data Source
AI summary
Techniques perform data access to a disk array. The disk array includes a parity disk and a plurality of data disks. Via such techniques, data is written to the parity disk in the disk array in response to a write request to a failed data disk in the disk array when the disk array is in a degraded state; and corresponding degraded storage position information is set in disk resource mapping information so as to indicate that the data are stored in the parity disk. Accordingly, enormous computing resources can be saved, and I/O operations required by reads in the degraded state can be reduced.