Flash Storage Stripe Recovery for Data Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash storage systems face limitations in data recovery due to the exhaustion of spare storage blocks and potential physical failures that prevent data recovery in failed flash storage devices.
Innovation Solution
A method that generates redundant data across multiple flash storage devices, allowing for data recovery by writing and reading data units and redundant data units at common physical addresses, and relocating them to new addresses when failures occur, ensuring uninterrupted operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If spare storage blocks are used to recover corrupt data in failed storage blocks, then data recovery capability is improved, but the number of available spare blocks is depleted, limiting long-term recovery capacity
Solution Approach 1:
The patent transitions from single-device spare block recovery to multi-device stripe-based recovery. By distributing data across multiple flash storage devices in stripes with redundancy, the system recovers data from any single device failure using corresponding stripes from other devices, effectively adding a spatial dimension to recovery capacity.
Solution Approach 2:
The system changes the recovery parameter from relying on finite spare blocks within a single device to using redundant stripes distributed across multiple devices. This parameter change transforms the recovery mechanism from depletion-based to renewable-based, where recovered data can be redistributed to maintain spare capacity.
2Device complexity
If data is stored in a single flash storage device, then storage simplicity is maintained, but physical failure of the device prevents data recovery
Solution Approach 1:
The patent segments data into multiple stripes distributed across different flash storage devices. Each stripe contains redundant information that can recover the original data. This segmentation allows the system to withstand physical failure of individual devices while maintaining data recovery capability.
Solution Approach 2:
The patent introduces redundant stripes as intermediaries between the original data and the failure event. These redundant stripes act as mediators that enable data recovery by providing alternative sources of information when a storage device physically fails.
3Reliability
If spare storage blocks are allocated for data recovery, then data loss prevention is improved, but the usable storage capacity is reduced
Solution Approach 1:
The patent makes storage blocks universal by enabling them to serve dual purposes: storing user data during normal operation and serving as recovery sources through stripe redundancy when failures occur. This multi-functionality eliminates the need for dedicated spare blocks, as any block can participate in recovery operations.
Solution Approach 2:
The system implements dynamic discarding and recovering of storage blocks through stripe redistribution. When data is recovered from a failed device, the recovered data and its redundant stripes are redistributed across available devices, continuously recovering usable capacity and making previously lost blocks available for future operations.
Data Source
AI summary
A data storage method, comprising, receiving host data to be written to a plurality of flash storage devices, allocating the host data to one or more data units of a plurality of data units, allocating pad data to one or more data units of the plurality of data units that have not been filled with host data and generating redundant data in a redundant data unit based on the plurality of data units. The method further comprises steps for writing the plurality of data units and the redundant data unit to a stripe across the plurality of flash storage devices, wherein each of the plurality of data units and the redundant data unit is written in the respective flash storage devices at a common physical address.