Multi-Threshold Read Commands for Analog Memory Cells
Solution Overview
Problem
Existing memory devices face inefficiencies in reading data from multi-level analog memory cells due to the need for multiple read thresholds and the overhead of memory access operations, particularly when re-reading bits or computing error correction metrics.
Innovation Solution
Implementing built-in multi-threshold read commands in memory devices that allow for efficient re-reading of data bits using a single instruction, with the option to adjust thresholds and compute soft decoding metrics, reducing memory access operations and improving data throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple read thresholds are used to read data from multi-level analog memory cells, then data accuracy is improved, but the number of memory access operations increases
Solution Approach 1:
The patent segments the reading process into two distinct phases: a first read operation using a first set of read thresholds, and a second read operation using a second set of read thresholds. This segmentation allows the system to perform different types of reads (initial read vs. re-read) with different threshold configurations, thereby improving data accuracy through multiple thresholds while managing memory access operations efficiently by only performing the second read when necessary
Solution Approach 2:
The patent implements dynamic threshold selection where the memory device can switch between different sets of read thresholds based on the reading requirements. The system dynamically determines whether to perform a first read or a second read based on conditions such as whether an error correction code can be successfully decoded, allowing flexible adaptation between using fewer thresholds (faster) and more thresholds (more accurate)
2Measurement precision
If re-reading operations are performed to compute error correction metrics, then error correction accuracy is improved, but data throughput requirements increase
Solution Approach 1:
The patent performs preliminary error correction by attempting to decode the error correction code after the first read operation. If the decoding succeeds, no second read is needed. The second read with different thresholds is performed only as a preliminary step when the initial error correction fails, allowing the system to compute error correction metrics more accurately without unnecessarily increasing data throughput requirements
Solution Approach 2:
The memory device performs self-service error correction by using its own built-in second read command to re-read data bits when needed for computing error correction metrics. This self-service mechanism allows the device to improve error correction accuracy using its own resources without requiring additional external read operations, thereby managing data throughput efficiently
Data Source
AI summary
A method for data storage includes storing data in a memory that includes multi-bit analog memory cells, each of which stores at least first and second data bits by assuming one of a predefined plurality of programming levels associated with respective storage values. The memory has at least a first built-in command for reading the first data bits of the memory cells by comparing the storage values of the memory cells to a first number of first thresholds, and a second built-in command for reading the second data bits of the memory cells by comparing the storage values of the memory cells to a second number of second thresholds, such that the first number is less than the second number. After storing the data, the first data bits are read from the memory cells by executing at least the second built-in command.