Storage Device Parity Group Garbage Collection for Drive Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems using the write-once data storage method, the increase in garbage data prolongs the data regeneration process during drive failures, potentially leading to extended redundancy reduction periods and increased data loss risks if another drive fails during recovery.
Innovation Solution
A storage device manages data across multiple logical hierarchies, executing a garbage collection process by copying valid data from one parity group to another, thereby shortening the drive recovery time and reducing data loss possibilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the write-once data storage method is used to manage data in storage drives, then data storage efficiency is improved, but the data regeneration time during drive failure is extended due to increased garbage data
Solution Approach 1:
The system performs preliminary garbage collection on the first logical area associated with the failure parity group before executing data regeneration. By proactively identifying and removing garbage data in advance, the system reduces the amount of unnecessary data that needs to be processed during the critical recovery phase, thereby shortening the overall data regeneration time while maintaining storage efficiency
Solution Approach 2:
The system divides the storage space into multiple logical areas associated with different parity groups. By segmenting the garbage collection process to target only the first logical area related to the failure parity group, rather than the entire storage system, the recovery process becomes more focused and efficient, reducing the time penalty associated with garbage data
2Quantity of substance
If garbage collection is performed by copying valid data within the same parity group, then storage space is preserved, but data loss risk increases if another drive fails during recovery
Solution Approach 1:
The system uses a second parity group as an intermediary destination for copying valid data during garbage collection. Instead of attempting to maintain data within the failing first parity group, the system transfers data to a healthy second parity group, creating a safe intermediate state. This intermediary approach preserves storage space while eliminating the risk of data loss from cascading failures within the same parity group
Data Source
AI summary
A storage device manages data stored in a storage drive in a plurality of logical hierarchies. The plurality of logical hierarchies include a writing hierarchy above a hierarchy of a parity group including a plurality of storage drives. The storage device writes received host data in a free area in the writing hierarchy. In the data recovery process for replacing a failure storage drive with a new storage drive, the storage device executes the garbage collection process on a first logical area in the writing hierarchy associated with the first parity group including the failure storage drive. In the garbage collection process, valid data is selected from the first logical area and copied to a second logical area associated with a second parity group different from the first parity group.