RAID Engine SSD Cell Failure Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current RAID controllers are not configured to manage solid state drives (SSDs) in read-only mode, which limits their functionality when cell failures exceed a certain threshold, leading to reduced user capacity and restricted access.
Innovation Solution
A RAID engine detects SSD cell failures exceeding a predetermined threshold, redirects writes to other drives in the RAID, and unmaps data stored on the SSD, effectively treating the SSD as isolated for writes while maintaining read access, thereby extending the time before the SSD enters read-only mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the SSD is configured with an over provisioned area for failover, then the reliability of the SSD is improved, but the user capacity is reduced
Solution Approach 1:
The over provisioned area serves as a pre-configured buffer or cushion that absorbs cell failures before they impact user capacity. By reserving 28% of total storage space as spare memory cells, the system prepares in advance to handle failures without immediately reducing accessible capacity, thus maintaining reliability while delaying the point at which user capacity is affected.
2Reliability
If the SSD enters read-only mode when cell failures exceed the over provisioned area, then the reliability of data integrity is improved, but the productivity and usability are reduced
Solution Approach 1:
The RAID controller acts as an intermediary between the SSD and the host system. When the SSD enters read-only mode due to excessive cell failures, the RAID controller intercepts write operations and redirects them to other healthy drives in the RAID array. This mediator approach allows the SSD to maintain data integrity by staying in read-only mode while the RAID system continues to accept write operations through alternative paths, thus preserving productivity.
Solution Approach 2:
The RAID controller is enhanced to perform multiple functions: it not only manages the RAID array but also actively monitors SSD cell failure counts and dynamically redirects write operations when the SSD enters read-only mode. This multi-functionality allows a single component to handle both normal RAID operations and failure mitigation, eliminating the need for separate mechanisms to handle read-only mode scenarios.
3Device complexity
If present RAID controllers do not address SSDs in read-only mode, then the device complexity is reduced, but the adaptability and versatility are reduced
Solution Approach 1:
The RAID controller implements dynamic behavior by continuously monitoring the number of failed cells on the SSD and adapting its operation accordingly. When failures remain within the over provisioned area, the controller operates normally. When failures exceed this threshold and the SSD enters read-only mode, the controller automatically changes behavior to redirect write operations. This dynamic adaptation allows the system to respond to changing conditions without requiring complex manual intervention or reconfiguration.
Data Source
AI summary
Managing a solid state drive (‘SSD’) in a Redundant Array of Inexpensive Drives (‘RAID’), including: detecting, by a RAID engine, a number of cell failures of the SSD exceeding a predetermined threshold; responsive to detecting the number of cell failures of the SSD exceeding the predetermined threshold: redirecting, by the RAID engine, a write of data not originally stored in the RAID to a drive in the RAID other than the SSD; and redirecting, by the RAID engine, writes originally directed to data stored on the SSD to another drive in the RAID, including unmapping the data stored on the SSD.