RAID Rebuilding Using Distributed Spare Logic Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAID systems face performance bottlenecks during disk rebuilding due to dedicated spare disks, which limit bandwidth and user IO performance, and increase the risk of data loss as all disk space is consumed, leading to inefficient resource utilization and prolonged response times.
Innovation Solution
The method involves determining a spare RAID group with spare capacity from a storage pool, building a spare logic unit, and using it to rebuild failed disks in a degraded RAID group, distributing write IO across all disks, thereby eliminating the need for dedicated spare disks and improving rebuilding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated spare disks are used for RAID rebuilding, then data redundancy is ensured, but rebuilding performance is limited and user IO performance deteriorates
Solution Approach 1:
The patent applies universality by enabling all disks in the storage pool to serve dual purposes: normal data storage and spare capacity for rebuilding. Instead of dedicating specific disks solely for spare capacity, every disk can contribute its unused space to the rebuilding process, allowing the system to dynamically allocate resources based on demand while maintaining data redundancy.
Solution Approach 2:
The patent merges the function of dedicated spare disks with normal data disks by creating a unified spare capacity pool. The spare capacity is distributed across multiple disks rather than concentrated on specific spare disks, combining resources from throughout the storage pool to improve rebuilding performance while maintaining the same level of data protection.
2Quantity of substance
If all disk space is consumed by RAID groups, then storage capacity is maximized, but rebuilding capability is compromised and data loss risk increases
Solution Approach 1:
The patent applies local quality by allowing different regions of the storage pool to have different functions at different times. Normal disks can dynamically allocate portions of their capacity as spare capacity when needed for rebuilding, while maintaining their primary data storage function. This localized flexibility allows the system to maintain full storage capacity while ensuring rebuilding capability is available when required.
3Reliability
If dedicated spare disks are allocated, then rebuilding can be performed, but effective disk utilization rate decreases and response time increases
Solution Approach 1:
The patent implements dynamics by making the spare capacity allocation flexible and adaptive rather than static. The system can dynamically adjust which disks provide spare capacity based on current workload, performance requirements, and failure scenarios. This dynamic allocation allows the system to optimize both rebuilding capability and normal operation efficiency, as disks can switch between providing user IO performance and providing spare capacity as needed.
Data Source
AI summary
Embodiments of the present disclosure provide a method and device for RAID rebuilding. In some embodiments, there is provided a computer-implemented method. The method comprises: determining a spare redundant array of independent disks (RAID) group with spare capacity from a plurality of disks included in at least one RAID group of a storage pool; building spare logic units from the spare RAID group; and in response to a RAID group of the at least one RAID group of the storage pool being in a degradation state, rebuilding a failed disk in a degraded RAID group using the spare logic units.