Memory ECC Re-Encoding After Failed Device Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor memory devices, the failure of a memory device reduces the size of the data unit, limiting the ability to identify additional failed devices due to reduced redundancy for error correction, especially when using symbol-based ECC codes.
Innovation Solution
The system re-encodes data units from symbol-based ECC to bit-based ECC after a memory device failure, allowing for higher individual bit error correction and enabling continued operation by re-encoding affected memory regions, even during the re-encoding process, and integrates this with wear leveling to manage data distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If symbol-based ECC codes are used for error correction, then correction capability for multiple failed devices is maintained, but upon memory device failure the reduced redundancy limits the ability to identify additional failed devices
Solution Approach 1:
The patent dynamically changes the ECC encoding parameters from symbol-based to bit-based when a memory device failure is detected. This parameter change allows the system to adapt the error correction mechanism to the current redundancy level, maintaining the ability to identify and correct errors even with reduced redundancy after device failure.
2Quantity of substance
If memory device failure occurs, then the data unit size is reduced, but this reduction limits continued error correction and failure detection capabilities
Solution Approach 1:
The patent implements a dynamic ECC re-encoding mechanism that activates upon detecting memory device failure. The system transitions from a static symbol-based ECC approach to a dynamic bit-based ECC approach, allowing the error correction capability to adapt to the reduced data unit size and maintain reliability throughout the device's operational life.
3Reliability
If re-encoding is performed after memory device failure, then error correction capability is enhanced, but system operation is interrupted during the re-encoding process
Solution Approach 1:
The patent performs re-encoding operations proactively during wear leveling cycles rather than waiting for errors to occur. By preliminarily re-encoding data with bit-based ECC during scheduled maintenance operations, the system enhances error correction capability without causing unexpected interruptions to normal system operation.
Data Source
AI summary
A technique includes determining that a given memory device of a plurality of memory devices has failed and in response to the determination that the given memory device has failed, re-encoding a data unit associated with the given memory device. The data unit is associated with a payload and a symbol-based error correction code. The re-encoding includes determining a bit-based error correction code for the payload and replacing the data unit in the memory with the payload and the bit-based error correction code.