Storage Data Block Allocation With Bit-Flip Qualification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage systems face issues with data loss due to unreliable storage blocks, as some memory cells may be damaged or have manufacturing defects, leading to decreased data resiliency and increased error correction operations.
Innovation Solution
An inline flash qualification process identifies blocks that can reliably store data by checking for bit flips and adding them to an active pool, preventing allocation of unreliable blocks, thus improving data resiliency and reducing error correction operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional storage systems allocate free blocks without qualification, then allocation speed is maintained, but data resiliency decreases due to unreliable blocks
Solution Approach 1:
The system performs preliminary qualification of free blocks by reading data and detecting bit flips before allocating them to storage operations. This advance screening ensures that only reliable blocks are added to the active free block pool, resolving the contradiction by prioritizing reliability checks before allocation occurs.
2Quantity of substance
If unreliable blocks are used for storage, then storage capacity is maximized, but error correction operations increase
Solution Approach 1:
The system converts the potentially harmful effect of bit flips in free blocks into a beneficial screening mechanism. By deliberately reading data into free blocks and detecting bit flips, the system identifies unreliable blocks and excludes them from the active pool, thereby reducing error correction operations while maintaining storage capacity.
3Reliability
If all free blocks are qualified by reading data, then data resiliency improves, but processing bandwidth is consumed
Solution Approach 1:
The system performs qualification only on free blocks that are candidates for allocation, rather than continuously qualifying all blocks in the storage system. This partial action approach qualifies blocks on-demand when they become available, improving data resiliency for allocated blocks while minimizing unnecessary processing bandwidth consumption.
Data Source
AI summary
Reading data stored at a free block of a storage device is read prior to allocating the free block for storage of data. A determination as to whether a number of bit flips of the data stored at the free block is below a threshold is made. The free block is added to a pool of active free blocks to be allocated for the storage of data upon determining that the number of bit flips of the data stored at the free block is below the threshold.


