Over-Sampling Read Memory Controller Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
NAND-based technology is costly while meeting the demand for high-density data storage, and existing magnetic storage technologies are less desirable due to weight and speed requirements.
Innovation Solution
Implementing over-sampling read (OSR) in memory devices to optimize controller performance by performing multiple internal reads within specific threshold ranges, reducing the need for multiple hard reads and enhancing error correction without decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple hard reads are performed to enhance error correction, then reliability is improved, but loss of time and use of energy increase
Solution Approach 1:
The patent performs a first read operation that reads more data than immediately necessary (excessive action), storing additional samples in a buffer. This partial over-read enables subsequent error correction operations to use this pre-fetched data without requiring additional read operations, thereby reducing total time while maintaining reliability
Solution Approach 2:
The patent performs read operations in advance of when the data is immediately needed, storing the extra samples in a buffer during the first read operation. This preliminary action ensures that when error correction is needed, the data is already available in memory, eliminating the need for time-consuming re-reads
2Reliability
If multiple hard reads are performed to enhance error correction, then reliability is improved, but use of energy increases
Solution Approach 1:
The patent performs a first read operation that reads more data than immediately necessary (excessive action), storing additional samples in a buffer. This partial over-read enables subsequent error correction operations to use this pre-fetched data without requiring additional read operations, thereby reducing total energy consumption while maintaining reliability
Solution Approach 2:
The patent maintains a continuous buffer of pre-read data that can be reused for multiple error correction operations. This continuity eliminates the need to repeatedly perform energy-consuming read operations, as the buffer provides a continuous supply of usable data for correction algorithms
3Reliability
If decoding is performed for error correction, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent extracts and uses only the necessary portions of the read data for error correction, storing extra samples in a buffer that can be selectively used. This extraction approach allows simple comparison-based error correction without requiring complex decoding algorithms, reducing device complexity while maintaining reliability
Solution Approach 2:
The patent uses simple buffer storage and direct data comparison methods instead of complex decoding algorithms. The buffer acts as a simple, low-cost mechanism that provides error correction capability without the complexity of traditional decoding, effectively using a simple disposable buffer rather than complex computational structures
Data Source
AI summary
A method of controller optimization utilizing over-sampling read (OSR) in a memory device includes performing a first internal read at a predetermined threshold level and transferring the first internal read measurement to the controller, performing a second internal read in a range that is between the predetermined threshold level plus a first predetermined value and the predetermined threshold level minus a second predetermined value, and determining whether a cell level falls in the range and transferring the second internal read measurement to the controller.