Compression Engine for Memory Codeword Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As page sizes in NAND devices decrease, the spare storage space for error-correcting code (ECC) is reduced, leading to a higher error rate and impacting the ability of memory sub-systems to correct errors. Additionally, market demand for larger logical block sizes is increasing, and conventional data compression approaches ignore ECC and metadata, limiting capacity and efficiency.
Innovation Solution
Incorporating a compression engine into the datapath of a memory sub-system controller, which compresses codewords including user data, ECC, and metadata for write operations and decompresses them for read operations, thereby improving code rate and error correction capabilities without requiring special NAND page layout changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If page sizes in NAND devices are decreased, then storage capacity increases, but spare storage space for ECC is reduced leading to higher error rates
Solution Approach 1:
The patent merges user data and ECC into a single codeword structure that is then compressed together. This integration allows the compressed representation to achieve higher code rates while maintaining error correction capability, effectively resolving the contradiction between increased storage capacity and maintained reliability in smaller page sizes.
Solution Approach 2:
The patent changes the parameter of code rate by implementing compression on the entire codeword (user data + ECC) rather than just user data. This parameter change enables the system to achieve code rates greater than 1, which directly addresses the limitation of having insufficient spare space for ECC in decreased page sizes.
2Quantity of substance
If conventional data compression approaches are used, then user data capacity increases, but ECC and metadata are ignored limiting overall efficiency
Solution Approach 1:
The patent combines user data, ECC, and metadata into a single compression unit (codeword). This merging ensures that all components benefit from compression simultaneously, improving overall data operation efficiency while maximizing user data capacity. The unified approach avoids the inefficiency of compressing only user data while leaving ECC and metadata uncompressed.
Solution Approach 2:
The compression engine is designed to universally compress any codeword structure containing user data, ECC, and metadata together. This multi-functional approach allows the same compression mechanism to improve both capacity and efficiency across different data types and error correction schemes, making the solution broadly applicable and highly efficient.
3Quantity of substance
If compression is applied only to user data, then user data storage efficiency improves, but overall code rate and error correction capabilities are limited
Solution Approach 1:
The patent merges user data and ECC into a unified compression target. By compressing the entire codeword together rather than user data alone, the system achieves higher overall compression ratios while preserving the full error correction capability. The compressed ECC maintains its error correction function while the entire structure achieves improved storage efficiency.
Data Source
AI summary
A processing device receives a request to write data to a memory device. The processing device generates a codeword based on the data. The codeword comprises the data and error correction code. The processing device generates a compressed codeword by compressing the codeword. The processing device stores the compressed codeword on a page of the memory device.