Logical Disk Virtualization for RAID Tray Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In RAID systems, removing and replacing a single disk triggers costly and resource-intensive data reconstruction across multiple RAID groups, especially when multiple disks on a single carrier are affected, leading to prolonged downtime and resource usage.
Innovation Solution
By configuring trays with multiple disks as a single logical disk, each disk is allocated to a different RAID group, allowing for rapid RAID recovery and minimizing the impact of disk removal on RAID groups, thus enabling quicker system recovery and reduced resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If multiple disks on a single carrier are treated as separate physical disks in RAID groups, then disk replacement flexibility is maintained, but data reconstruction time and processing resources increase significantly
Solution Approach 1:
The patent merges multiple physical disks located on the same carrier into a single logical disk unit. This logical disk is then allocated to a single RAID group, so that when one or more disks on the carrier fail, the entire carrier is treated as one failure unit rather than multiple separate failures. This consolidation reduces the number of RAID groups that need to perform reconstruction operations simultaneously, thereby reducing overall reconstruction time and processing resource consumption.
2Loss of energy
If each disk on a carrier is allocated to different RAID groups, then disk replacement flexibility is maintained, but system resources are wasted during reconstruction of multiple terabyte volumes
Solution Approach 1:
The patent consolidates multiple disks on a carrier into a single logical disk that is allocated to one RAID group. This ensures that when a carrier is removed or a disk fails, only one RAID group is affected rather than multiple RAID groups having their components on the same carrier. The single RAID group performs reconstruction using its parity disk, consuming processing resources for only one volume reconstruction operation at a time, thereby reducing overall energy and processing resource consumption.
3Productivity
If a tray with multiple disks is removed for replacement, then all disks on the tray are replaced simultaneously, but this causes multiple RAID groups to be affected and increases reconstruction cost
Solution Approach 1:
The patent merges all disks on a carrier into a single logical disk allocated to one RAID group. When a carrier is removed, the system identifies which RAID groups are affected and performs reconstruction only for those groups. Because all disks on the carrier share the same RAID group allocation, the system can efficiently manage reconstruction operations and replace carriers faster without triggering multiple separate reconstruction processes, thereby increasing system recovery speed while controlling reconstruction time.
Data Source
AI summary
A computer storage system includes multiple disk trays, each disk tray holding two or more physical disks. The disks on a single tray are virtualized into a single logical disk. The single logical disk reports to the RAID (redundant array of inexpensive disks) subsystem, creating the impression that there is one large capacity disk. In one implementation, each disk in the tray is allocated to a different RAID group. By allocating the disks in a tray to different RAID groups, if the tray is removed, only a portion of several different RAID groups are removed. This arrangement permits a simple reconstruction of the RAID groups if a disk tray is removed from the system.