Memory Block Stuck Bit Handling via Generator Matrix Submatrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Charged-based memory storage technologies, such as DRAM, face scalability issues due to increasing process variation, and phase change memory (PCM) has a limited lifetime as cells become inoperable over time, leading to data loss.
Innovation Solution
A code generator matrix is used to create codewords from write operations, where some bits are designated as 'stuck' to extend the operational life of writable bits, with pre-computed submatrices stored in ROM to enhance write operation performance, allowing correction of up to d−1 stuck bits without revealing their locations during read operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If conventional write operations are used without designating stuck bits, then all bits can be written initially, but the operational life of the memory block is limited as bits become inoperable over time
Solution Approach 1:
The system performs preliminary identification of stuck bits before write operations and pre-computes submatrices of the generator matrix corresponding to non-stuck bits. This advance preparation allows the memory system to adaptively handle degraded memory blocks, extending their operational life by writing only to functional bits while maintaining error correction capability.
Solution Approach 2:
The generator matrix is partitioned into submatrices based on the specific stuck bit pattern of each memory block. This local adaptation allows the system to tailor the encoding scheme to the actual functional state of each block, maximizing the usable capacity while ensuring reliability through targeted error correction on non-stuck bits only.
2Productivity
If the entire generator matrix is used to generate codewords, then complete error correction is achieved, but write operation performance is reduced due to unnecessary computation on stuck bits
Solution Approach 1:
The system extracts and removes the columns corresponding to stuck bits from the generator matrix, creating a reduced submatrix that contains only the columns for non-stuck bits. This extraction eliminates unnecessary computation on inoperable bits while preserving the error correction capability for the functional bits, thereby improving write performance without sacrificing reliability.
Solution Approach 2:
Instead of applying the full generator matrix to all bits, the system applies a partial generator matrix (submatrix) only to the non-stuck bits. This partial action is sufficient to provide error correction for the usable portion of the memory block, achieving the necessary reliability while significantly reducing computational overhead and improving write performance.
3Reliability
If stuck bit locations are revealed during read operations, then error correction can be optimized, but security and data integrity are compromised
Solution Approach 1:
The system uses a copy of the generator matrix structure (submatrices) stored in ROM that corresponds to the stuck bit pattern, but does not expose the actual stuck bit locations during read operations. The pre-computed submatrices enable error correction optimization without revealing sensitive information about which bits are inoperable, maintaining both reliability and information security.
Data Source
AI summary
A generator matrix is provided to generate codewords from messages of write operations. Rather than generate a codeword using the entire generator matrix, some number of bits of the codeword are determined to be, or designated as, stuck bits. One or more submatrices of the generator matrix are determined based on the columns of the generator matrix that correspond to the stuck bits. The submatrices are used to generate the codeword from the message, and only the bits of the codeword that are not the stuck bits are written to a memory block. By designating one or more bits as stuck bits, the operating life of the bits is increased. Some of the submatrices of the generator matrix may be pre-computed for different stuck bit combinations. The pre-computed submatrices may be used to generate the codewords, thereby increasing the performance of write operations.