Spare Disk Array for RAID Rebuilding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In traditional storage systems, the rebuilding process of a faulty disk often becomes bottlenecked due to high writing bandwidth requirements on a single spare disk, and concurrent random I/O requests can slow down or interrupt the rebuilding process, potentially leading to data loss if another disk fails.
Innovation Solution
The solution involves creating a spare disk array using SSDs with lower Write Per Day (WPD) to distribute rebuilding operations across multiple disks, reducing the number of spare disks required and improving utilization efficiency by reallocating spare logic storage units on demand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a single spare disk is used for rebuilding a faulty disk in a RAID group, then the rebuilding process can be completed, but the writing bandwidth of the spare disk becomes a bottleneck for the whole rebuilding process
Solution Approach 1:
The patent divides the single spare disk into multiple spare disks, creating a spare disk array. The rebuilding workload is segmented and distributed across multiple spare disks simultaneously, thereby increasing the aggregate writing bandwidth and accelerating the rebuilding process without requiring excessive spare disk resources.
Solution Approach 2:
The patent merges multiple spare disks into a unified spare disk array that works cooperatively. Multiple disk array groups can share the same spare disk array, combining their rebuilding operations to utilize the collective capacity of the spare disks more efficiently.
2Ease of operation
If the storage system receives random I/O requests during the rebuilding process, then the storage system remains operational, but the response time of random I/O requests is affected and the rebuilding process may slow down or be interrupted
Solution Approach 1:
The patent segments the spare disk capacity into multiple independent spare logic storage units that can be dynamically allocated. This segmentation allows the system to maintain operational availability by serving I/O requests while preserving sufficient capacity and bandwidth for the rebuilding process, reducing the risk of interruption.
3Productivity
If a large number of random I/O requests are received during rebuilding, then the storage system handles user requests, but the rebuilding process may slow down or be interrupted, potentially causing data loss
Solution Approach 1:
The patent implements dynamic allocation of spare logic storage units based on real-time system conditions. The system can flexibly adjust the capacity and bandwidth allocated to rebuilding operations versus I/O requests, ensuring data safety while maintaining productive I/O handling capacity under varying load conditions.
Data Source
AI summary
Embodiments of the present disclosure relate to a method, device and a computer program product for managing a storage system. The storage system includes a disk array. A method includes determining, based on a first number of disks in the disk array, a second number of spare disks for the disk array. The method further includes creating a spare disk array with the second number of spare disks. The method further includes, in response to a first disk in the disk array failing, allocating a spare logic storage unit from the spare disk array for rebuilding the first disk. In addition, the method further includes rebuilding the first disk with the spare logic storage unit.