DRAM Weak Cell Encoding for Error Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic random access memory (DRAM) cells exhibit varying charge retention rates due to manufacturing variations, leading to weak cells that require frequent refresh, consuming resources and causing errors if not properly managed.
Innovation Solution
A method is introduced to determine the location of weak cells in a DRAM row and encode data into a bit vector, ensuring compatibility with the reliable logic state of weak cells, allowing for error-free data writing and decoding without accessing weak cell information, thereby optimizing refresh rates and reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If weak cells are identified and managed through encoding, then data reliability is improved, but device complexity increases due to additional encoding logic and bit vector manipulation
Solution Approach 1:
The system performs preliminary identification of weak cells during manufacturing or initialization, storing their locations in a lookup table. This advance preparation allows the encoding logic to simply reference pre-identified weak cell positions rather than dynamically analyzing each cell, reducing operational complexity while maintaining reliability
Solution Approach 2:
The encoding scheme applies different handling to different bit positions based on whether they correspond to weak or strong cells. By locally adapting the encoding/decoding operation at specific positions (inverting bits for weak cell positions), the system achieves reliable data storage without requiring complete system redesign, thus managing complexity through localized modifications
2Reliability
If refresh operations are performed more frequently to accommodate weak cells, then data retention reliability is improved, but power consumption increases due to additional refresh cycles
Solution Approach 1:
Weak cell locations are determined and stored in advance during manufacturing or initialization. This preliminary identification allows the system to implement encoding that compensates for weak cell charge loss without requiring increased refresh rates, as the encoding already accounts for the weak cell behavior pattern
Solution Approach 2:
The patent converts the harmful effect of weak cells (charge loss) into a beneficial pattern by using the known weak cell positions to drive the encoding scheme. The weak cell locations become the basis for a systematic encoding/decoding approach that actually improves data reliability while allowing standard refresh rates to be maintained
3Measurement precision
If the bit vector is extended to include weak cell information, then data accuracy is improved, but the amount of data to be processed increases
Solution Approach 1:
Weak cell location information is determined and stored in advance in a separate lookup table or configuration structure. This preliminary storage means the bit vector itself doesn't need to expand to include weak cell information, as that information is already available from the pre-established lookup table, thus maintaining data volume efficiency while ensuring accuracy
Solution Approach 2:
The system segments information storage into two parts: the lookup table containing weak cell location data and the bit vector containing actual data to be stored. This segmentation allows the bit vector to remain compact for data storage while the lookup table handles the metadata about weak cells, avoiding unnecessary expansion of the data vector itself
Data Source
AI summary
This disclosure includes a method for preventing errors in a DRAM (dynamic random access memory) due to weak cells that includes determining the location of a weak cell in a DRAM row, receiving data to write to the DRAM, and encoding the data into a bit vector to be written to memory. For each weak cell location, the corresponding bit from the bit vector is equal to the reliable logic state of the weak cell and the bit vector is longer than the data.