Dynamic RAID Stripe Grouping for Solid State Memory Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional RAID configurations in data storage devices rely on fixed geometry mapping, which cannot adjust to localized reliability issues in storage elements, leading to potential data loss due to uneven sector reliability across stripes.
Innovation Solution
The use of high-granularity reliability information from individual storage elements in solid-state storage devices to dynamically group and re-distribute elements within RAID stripes, ensuring a homogeneous reliability metric across the storage device, with a reliability module that monitors conditions like erase counts and bit errors to adjust the redundancy configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed geometry mapping is used in traditional RAID configurations, then the storage system structure is simple and easy to implement, but the reliability is reduced due to inability to adjust to localized reliability issues
Solution Approach 1:
The patent implements dynamic stripe group formation by grouping storage elements based on their reliability metrics rather than using fixed geometry mapping. The system periodically monitors reliability metrics and reconfigures stripe groups to include only reliable storage elements, transforming a static RAID configuration into a dynamic one that adapts to changing reliability conditions.
Solution Approach 2:
The patent applies local quality by treating different storage elements differently based on their individual reliability metrics. Instead of uniform treatment, the system identifies and groups only the most reliable storage elements into active stripe groups, while isolating or removing unreliable elements, thereby optimizing reliability locally within each stripe group.
2Reliability
If storage elements with low reliability are included in RAID stripes, then the storage capacity is maximized, but the risk of data loss increases due to uneven sector reliability
Solution Approach 1:
The system monitors and evaluates the reliability quality of individual storage elements and applies different treatment based on their quality level. High-reliability elements are included in active stripe groups while low-reliability elements are excluded or placed in separate groups, ensuring that only quality-assured storage elements contribute to data protection.
Solution Approach 2:
The patent extracts or removes storage elements with unreliable characteristics from the active RAID stripe groups. By identifying and separating out problematic storage elements, the system maintains the integrity and reliability of the remaining active stripe groups while preserving the overall storage capacity through selective inclusion of reliable elements.
3Reliability
If redundancy configuration is fixed, then the system is simple to manage, but it cannot prevent data loss when localized storage elements fail
Solution Approach 1:
The system implements feedback mechanisms by periodically monitoring reliability metrics of storage elements and using this information to dynamically adjust stripe group configurations. This closed-loop approach ensures that the redundancy configuration automatically adapts to changing reliability conditions without requiring manual intervention.
Solution Approach 2:
The redundancy configuration system performs self-adjustment based on monitored reliability metrics. The system automatically identifies unreliable storage elements and reconfigures stripe groups without external intervention, enabling the RAID system to self-optimize its reliability configuration over time.
Data Source
AI summary
Embodiments of the present invention use high granularity reliability information (e.g., from individual pages, blocks, etc.) in a solid state storage device to vary the number of elements in each RAID stripe and to combine the elements in a stripe to achieve a more homogenous reliability metric across the device. In one embodiment, a reliability metric of a stripe group of storage elements is calculated based on monitored conditions of the storage elements such as erase counts, number of bit errors encountered, calculated voltage reference values, etc. The reliability metrics of the stripe groups are used to decide how many storage elements and which storage elements should be combined in the redundant RAID stripes to achieve a desired probability of data loss for the overall device. The target error probability could be fixed for the life of the storage device or adjusted as the device wide error rates increase.