Memory Error Correction Flip-Bit Routing for Lower Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Error correction in memory devices introduces significant latency due to indirect routing of bits, where data bits must wait for error-corrected flip bits to be propagated through the system, affecting access times.

Innovation Solution

Transmitting error-corrected flip bits directly to the flip decision component on a dedicated line, bypassing the need for data bits to wait for error correction, allowing data to be transmitted without delay.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If error correction is performed using indirect routing of bits through data bit error correction components, then error correction functionality is achieved, but latency is significantly increased due to propagation delays

Engineering Contradiction:
Improveerror correctionVSAvoidlatency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the error correction path by creating a dedicated flip bit error correction component that operates independently from the data bit error correction components. This segmentation allows flip bits to be corrected separately and directly routed to the flip decision component, bypassing the indirect path through data bit error correction components that causes propagation delays.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a dedicated error correction path as an intermediary mechanism between the flip bit storage and the flip decision component. This dedicated path acts as a mediator that directly transmits error-corrected flip bits without requiring them to traverse through the data bit error correction components, thereby eliminating the propagation delays associated with indirect routing.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If data bits wait for error-corrected flip bits to be propagated through the system, then error correction accuracy is maintained, but access times are affected

Engineering Contradiction:
Improveerror correction accuracyVSAvoidaccess time
Core Design Contradiction:
ReliabilityVSSpeed

Solution Approach 1:

The patent implements preliminary action by correcting flip bits in advance through a dedicated error correction path before they are needed for data bit correction decisions. The flip bit error correction component processes flip bits independently and directly provides corrected values to the flip decision component, so that when data bits need error correction, the corrected flip bits are already available, eliminating waiting time.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If indirect routing is used for error correction, then comprehensive error correction is achieved, but system efficiency is reduced

Engineering Contradiction:
Improveerror correction coverageVSAvoidsystem efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the error correction system into independent functional units: a dedicated flip bit error correction component, a flip decision component, and data bit error correction components. This segmentation allows each component to operate independently and in parallel, improving overall system efficiency while maintaining comprehensive error correction coverage through the coordinated operation of these specialized units.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS10951232B2Error correction bit flipping scheme
Publication Date: 2021.03.16 MICRON TECHNOLOGY INC
  • US10951232B2 patent drawing
  • US10951232B2 patent drawing
  • US10951232B2 patent drawing

AI summary

Methods, systems, and devices for operating a memory device are described. An error correction bit flipping scheme may include methods, systems, and devices for performing error correction of one or more bits (e.g., a flip bit) and for efficiently communicating error correction information. The data bits and the flip bit (e.g., an error corrected flip bit) may be directly transmitted (e.g., to a flip decision component). The flip bit may be transmitted to the flip decision component over a dedicated and/or unidirectional line that is different from one or more other lines that carry data bits (e.g., to the flip decision component).