Flash Memory Seed Calculation for Faster Random Data Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Flash memory devices face challenges in performing random read operations efficiently due to the need for sequence generators, which can cause delays and require significant storage space for seeds, leading to variations in threshold voltages and reduced read margins.
Innovation Solution
A method and circuit for randomizing and de-randomizing data in flash memory devices using finite field arithmetic operations to generate seeds without driving a sequence generator, allowing for rapid seed calculation and de-randomization, thereby improving read performance and reducing threshold voltage variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a sequence generator is used to generate seeds for random read operations, then random data can be read from flash memory, but the read speed is reduced due to generation delays and significant storage space is consumed for seed storage
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing seed values in a lookup table during manufacturing or initialization. When a random read operation is requested, the system directly retrieves the pre-computed seed from the lookup table based on the requested location, eliminating the need for real-time sequence generation and significantly reducing read delays.
Solution Approach 2:
The patent uses copying by creating a lookup table that contains pre-generated seed values corresponding to different memory locations. Instead of generating seeds on-demand during read operations, the system copies the appropriate seed from the lookup table, which is a much faster operation that improves read speed while minimizing time loss.
2Adaptability or versatility
If a sequence generator is used to generate seeds, then random read operations can be performed, but significant storage space is required for seed storage leading to increased device complexity
Solution Approach 1:
The patent extracts the seed generation function from the main read operation by using a separate lookup table structure. The lookup table stores only the essential seed values needed for random reads, separating this functionality from the sequence generator hardware and reducing the overall device complexity while maintaining random read capability.
Solution Approach 2:
The patent changes the parameter of seed storage from dynamic generation to static pre-computed values. By transforming seeds into fixed lookup table entries indexed by memory location, the system reduces the complexity of seed management while preserving the adaptability for random read operations across different memory locations.
3Ease of operation
If sequence generators are used for seed generation, then random data access is enabled, but threshold voltage variations increase reducing read margins
Solution Approach 1:
The patent applies preliminary action by pre-calculating seed values that are optimized to minimize threshold voltage variations. These pre-computed seeds in the lookup table are designed to distribute read operations more uniformly across memory cells, reducing the impact of threshold voltage variations and improving read margins while maintaining ease of random data access.
Data Source
AI summary
A random data reading method of a nonvolatile memory device includes receiving an initial seed corresponding to a selected page of the nonvolatile memory device and relative location information of read-requested random data in the selected page. The method further includes generating a seed for randomizing the random data by subjecting the initial seed and the location information to a finite field arithmetic operation, and de-randomizing the random data based on a random sequence generated from the seed.