Memory ECC Pipelining for Faster Write Mask Operations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

As semiconductor memory sizes shrink and voltage levels decrease, memory devices become more prone to errors due to reduced charge margins and weak cells, leading to increased soft errors and variable refresh times, which can be exacerbated by ECCs that increase operation time, thereby reducing memory performance.

Innovation Solution

Implementing pipelined error correction coding (ECC) by concurrently executing multiple memory operations, including read and write mask operations, to reduce the overall execution time of write mask operations and enhance memory performance while maintaining data integrity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If error correction code (ECC) is used to correct data errors in memory cells, then data integrity is improved, but the time required for memory operations increases

Engineering Contradiction:
Improvedata integrityVSAvoidmemory operation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent pre-charges local input-output lines before actual data transfer operations. This preliminary action prepares the circuit paths in advance, reducing the overall operation time when ECC processing is performed, thereby mitigating the time penalty associated with error correction

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements variable refresh times for memory cells based on their individual charge leakage characteristics. By periodically refreshing weak cells more frequently and strong cells less frequently, the system maintains data integrity while optimizing the timing of ECC operations to minimize performance impact

Inventive Principle:
Principle #19Periodic action

2Area of stationary object

If feature sizes of semiconductor memory are reduced to produce smaller memory components, then memory capacity per area is improved, but the margin of error for detecting data state decreases

Engineering Contradiction:
Improvememory capacity per areaVSAvoiddata state detection margin
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent performs preliminary charging of memory cells with optimized voltage levels and durations tailored to each cell's characteristics. This preliminary action ensures that even in reduced-feature-size memory cells, sufficient charge margin is established before read operations, improving detection precision despite smaller cell sizes

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent dynamically adjusts charging parameters such as voltage level, charging duration, and refresh timing based on individual cell characteristics and environmental conditions. This parameter optimization compensates for reduced charge margins in smaller memory cells, maintaining reliable data state detection

Inventive Principle:
Principle #35Parameter changes

3Use of energy by stationary object

If voltage levels provided to semiconductor memory are reduced to limit power consumption, then power consumption is reduced, but memory cells become more prone to soft errors

Engineering Contradiction:
Improvepower consumptionVSAvoidsusceptibility to soft errors
Core Design Contradiction:
Use of energy by stationary objectVSReliability

Solution Approach 1:

The patent optimizes voltage levels and charging durations based on individual cell characteristics and operational conditions. By dynamically adjusting these parameters, the system maintains sufficient charge margins for reliable operation at lower voltage levels, reducing power consumption while mitigating susceptibility to soft errors and alpha particle interactions

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12500610B2Apparatuses and methods for pipelining memory operations with error correction coding
Publication Date: 2025.12.16 LODESTAR LICENSING GROUP LLC
  • US12500610B2 patent drawing
  • US12500610B2 patent drawing
  • US12500610B2 patent drawing

AI summary

Apparatuses and methods for pipelining memory operations with error correction coding are disclosed. A method for pipelining consecutive write mask operations is disclosed wherein a second read operation of a second write mask operation occurs during error correction code calculation of a first write mask operation. The method may further including writing data from the first write mask operation during the error correction code calculation of the second write mask operation. A method for pipelining consecutive operations is disclosed where a first read operation may be cancelled if the first operation is not a write mask operation. An apparatus including a memory having separate global read and write input-output lines is disclosed.