Memory Controller Permanent Bit Error Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing memory systems face challenges in handling permanent bit errors, particularly in high-radiation environments, where recycling power is necessary to correct single event latchup (SEL) errors, leading to system downtime and increased costs due to the need for specialized circuitry and more expensive radiation-insulated memory devices.

Innovation Solution

A memory system with a primary and secondary storage partition, along with a memory controller that detects and corrects bit errors using an error detection and correction (EDAC) circuit, flags permanent errors, and redirects subsequent operations to the secondary partition, eliminating the need for power recycling by distinguishing between non-permanent and permanent errors based on previous access history.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If power recycling is used to correct single event latchup (SEL) errors, then permanent bit errors can be corrected, but system downtime occurs and specialized circuitry is required

Engineering Contradiction:
Improveerror correction capabilityVSAvoidsystem downtime
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The memory device is divided into multiple banks, allowing the system to continue operating with unaffected banks while one bank undergoes correction or replacement, thereby eliminating system downtime associated with error correction

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a replacement mechanism where affected memory banks are replaced with spare banks, copying the functional capability without requiring power recycling or causing system downtime

Inventive Principle:
Principle #26Copying

2Reliability

If radiation-insulated memory devices are used to prevent permanent bit errors, then reliability in high-radiation environments is improved, but cost increases

Engineering Contradiction:
Improveresistance to permanent bit errorsVSAvoidcost
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

The patent uses standard, cost-effective memory devices with built-in error detection and correction capabilities, replacing them when permanent errors occur rather than using expensive radiation-insulated memory devices

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The memory controller automatically detects errors, identifies affected banks, and performs replacement operations without external intervention, maintaining reliability while using standard components

Inventive Principle:
Principle #25Self-service

3Reliability

If specialized circuitry is added to power supply to enable power recycling, then permanent bit errors can be corrected, but device complexity increases

Engineering Contradiction:
Improvepermanent error handlingVSAvoidcircuitry requirements
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts the error correction function from the power supply circuitry and relocates it to the memory controller, which uses error detection codes to identify and handle permanent errors without requiring specialized power recycling circuitry

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the mechanical/power-based error correction method (power recycling) with a control-based method using error detection and correction codes, eliminating the need for specialized power supply circuitry

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP2423815B1System for handling of permanent bit errors in memory devices
Publication Date: 2019.05.15 HAMILTON SUNDSTRAND CORP
  • EP2423815B1 patent drawingFigure 1
  • EP2423815B1 patent drawingFigure 2A
  • EP2423815B1 patent drawingFigure 2B

AI summary

A memory system (10) including a primary memory storage partition (17), a secondary memory storage partition (18), and a memory controller (14) that is connected to read and write to the primary memory storage partition (17) and detects a permanent bit error at an address associated with the primary memory storage partition (17). In response to a detected permanent bit error, the memory controller (14) stores data from the address associated with the permanent bit error to an address associated with the secondary memory storage partition (18).