Storage Control Parity Redundancy for RAID Failure Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face challenges in maintaining data redundancy and access speed when a storage device fails, particularly when rebuilding data across RAID groups without a hot spare, leading to reduced processing speed and increased burden during data rebuilding and rearrangement.
Innovation Solution
A storage system with a storage control apparatus that manages redundancy by distributing parity data across multiple storage devices, allowing for the generation and storage of rebuilt data in parity areas of another logical storage area, thereby reducing the need for data rearrangement and maintaining access speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If one of the storage devices belonging to the second RAID group is separated from the second RAID group and rebuilding is performed by using the separated storage device, then data redundancy is restored, but the use of the second RAID group is resumed after the RAID level is changed and the data redundancy is lowered, and the processing for rebuilding and rearrangement of the data may affect the processing for access to the storage devices in response to I/O requests from a host apparatus, and lower the speed of the processing for access
Solution Approach 1:
The patent segments the storage system into multiple logical storage areas (first logical storage area and second logical storage area), each with its own parity data distribution. When failure occurs in the first logical storage area, rebuilding is performed using parity storage areas in the second logical storage area independently, without affecting the data arrangement and access processing in the second logical storage area. This segmentation isolates the rebuilding impact to specific regions, maintaining overall system access speed.
Solution Approach 2:
The patent uses parity storage areas as intermediary regions for data rebuilding. Instead of directly rearranging data in the affected RAID group, the system writes rebuilt data to parity storage areas in another logical storage area, which then serves as the source for restoring the failed storage devices. This intermediary approach avoids direct data movement in the affected area, reducing access overhead.
2Reliability
If rebuilding is performed by separating a storage device from the second RAID group, then data redundancy is restored, but data are rebuilt and rearranged when the RAID level is changed, which increases the processing burden during rebuilding
Solution Approach 1:
The patent divides the storage system into separate logical storage areas with independent parity data management. When rebuilding is needed in the first logical storage area, only the parity storage areas in the second logical storage area are utilized, leaving the data arrangement in the second logical storage area unchanged. This segmentation eliminates the need for comprehensive data rearrangement across the entire system, reducing processing complexity.
Solution Approach 2:
Instead of performing full data rearrangement across the entire RAID group when changing RAID levels, the patent applies partial action by only modifying the necessary parity storage areas in the second logical storage area. The data in the second logical storage area that is not involved in the rebuilding process remains untouched, reducing the overall processing burden while still achieving the required data redundancy restoration.
Data Source
AI summary
A storage system including first storage devices constituting a first logical storage area, second storage devices constituting a second logical storage area; and a storage control apparatus. The storage control apparatus manages the first and second logical storage areas so that the data stored in the first and second logical storage areas have redundancy, and parity data for the data stored in the second logical storage area are stored in parity storage areas arranged in part of the second storage devices. When part of the first storage devices constituting part of the first logical storage area fail, the storage control apparatus generates part of the data stored, before the failure, in the part of the first storage devices, and stores the generated part of the data in at least part of the second parity storage areas in the second logical storage area.