NAND Flash Page Buffer Throughput via Bit Line Select Gates

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The limitations of traditional NAND flash memory architectures, where the number of page buffers is limited due to their large size, restricting read/write performance and throughput, as they occupy a significant portion of the memory die.

Innovation Solution

The implementation of bit line select gates that connect page buffers to multiple bit lines, allowing for simultaneous programming and reading of multiple bit lines, along with additional pass gates and data registers to enhance operation, and novel programming and reading operations to increase performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the number of page buffers is increased to improve read/write throughput, then data read/write throughput is improved, but the area occupied by page buffers increases significantly

Engineering Contradiction:
Improvedata read/write throughputVSAvoidarea occupied by page buffers
Core Design Contradiction:
ProductivityVSArea of stationary object

Solution Approach 1:

The patent divides the memory array into multiple sub-arrays, with each sub-array having its own dedicated page buffer. This segmentation allows parallel access to multiple sub-arrays simultaneously, effectively increasing throughput without requiring a single large page buffer that would occupy excessive die area. Each sub-array can be independently accessed through its own page buffer, enabling concurrent operations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces bit line select gates that enable a single page buffer to connect to and control multiple bit lines across different sub-arrays. This adds a dimensional aspect to the page buffer connectivity, allowing one page buffer to serve multiple sub-arrays by selectively enabling connections to different bit lines, thereby increasing throughput without proportionally increasing page buffer area.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Area of stationary object

If traditional NAND flash memory architecture is used with limited page buffers, then die area is reduced, but read/write performance and throughput are restricted

Engineering Contradiction:
Improvedie areaVSAvoidread/write performance
Core Design Contradiction:
Area of stationary objectVSProductivity

Solution Approach 1:

The patent makes page buffers multi-functional by enabling each page buffer to serve multiple sub-arrays through bit line select gates. A single page buffer can perform read and write operations on multiple sub-arrays by selectively activating connections to different bit lines, effectively increasing performance without requiring proportional increases in page buffer quantity or area.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces dynamically controllable bit line select gates that can selectively connect page buffers to different bit lines and sub-arrays based on operational requirements. This dynamic connectivity allows the same hardware resources to be flexibly allocated to different sub-arrays, optimizing performance while maintaining efficient use of die area.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS12142329B2Methods and apparatus for NAND flash memory
Publication Date: 2024.11.12 NEO SEMICON INC
  • US12142329B2 patent drawing
  • US12142329B2 patent drawing
  • US12142329B2 patent drawing

AI summary

Methods and apparatus for memory operations disclosed. In an embodiment, a method is provided for programming multiple-level-cells. The method includes programming data to single-level-cells (SLC) on SLC word lines using SLC programming operations, applying ramp data to the SLC word lines to determine selected ramp data that matches the data stored in (SLC) cells, and programming multiple-level cells to have a voltage threshold level that is associated with the ramp data. In an embodiment, an apparatus includes a first plane having a plurality of first cell strings coupled to a first page buffer. Each first cell string comprises a plurality of multiple-level cells. The apparatus also includes a second plane having a plurality of second cell strings coupled to a second page buffer. Each second cell string comprises a plurality of single-level cells. The apparatus is also configured so that the first page buffer is connected to communicate with the second page buffer.