In-Field Memory Reliability Assessment via Host Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods lack the ability to proactively assess and mitigate reliability issues in computer memory arrays in the field, leading to potential significant errors that can adversely affect system operation.

Innovation Solution

A computer-implemented method and system that takes control of a portion of a computer memory circuit, utilizes a memory bus, moves data to a storage device, executes logical operations to assess reliability, and transmits assessment data to a memory controller, allowing for proactive identification and mitigation of reliability issues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional memory operation methods are used, then system operation continues without interruption, but reliability issues in memory arrays cannot be proactively assessed or mitigated

Engineering Contradiction:
Improvememory reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by periodically taking control of memory circuits during idle periods to perform reliability assessments before failures occur. The host system proactively executes assessment routines that evaluate memory cell health, detect potential issues, and mitigate problems before they cause system errors, thus improving reliability without requiring fundamental changes to the memory architecture.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary assessment mechanism that acts as a mediator between the host system and memory circuits. During idle periods, the host takes control of the memory bus and executes assessment routines that indirectly evaluate memory reliability without requiring direct intervention in the memory control logic, thereby maintaining system simplicity while enabling proactive reliability monitoring.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If memory reliability assessment is implemented, then potential errors can be identified and mitigated, but system resources and time are consumed during assessment operations

Engineering Contradiction:
Improvememory reliabilityVSAvoidassessment time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies periodic action by scheduling memory reliability assessments during idle periods rather than continuously. The host system monitors for idle conditions and performs assessments at appropriate intervals, balancing the need for reliability evaluation with the requirement to minimize impact on system operational time. This periodic approach ensures timely detection of memory issues without consuming excessive resources during active operations.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent implements partial action by assessing only portions of the memory array during each idle period rather than the entire memory space. The host system selectively evaluates specific memory circuits or regions based on priority and resource availability, performing sufficient assessment to detect potential failures without the overhead of complete memory scanning, thus reducing time loss while maintaining effective reliability monitoring.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If control of memory circuit is taken for assessment, then reliability evaluation can be performed, but normal memory operations are interrupted

Engineering Contradiction:
Improvememory reliabilityVSAvoidmemory operation throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent resolves this contradiction by implementing periodic assessments during idle periods when the memory system is not actively being accessed. The host system monitors for idle conditions and only takes control of the memory circuit when no operations are pending, ensuring that reliability evaluations are performed without interrupting normal memory throughput during active operations.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent applies preliminary action by performing reliability assessments in advance during idle periods before memory operations resume. This proactive approach ensures that memory reliability is evaluated and potential issues are mitigated before they can impact normal operations, maintaining high productivity during active periods while still achieving thorough reliability monitoring.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10255986B2Assessing in-field reliability of computer memories
Publication Date: 2019.04.09 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10255986B2 patent drawing
  • US10255986B2 patent drawing
  • US10255986B2 patent drawing

AI summary

The present invention provides a computer implemented method, system, and computer program product of assessing in-field reliability of computer memories. In an embodiment, the present invention includes taking control of a portion of a computer memory circuit, utilizing a portion of a computer memory bus associated with the portion of the computer memory circuit, moving computer memory circuit data stored in the portion of the computer memory circuit to a host computer storage device, executing a set of logical operations assessing reliability of the portion of the computer memory circuit, resulting in assessment data stored in a reliability error monitor (REM) computer storage device, transmitting the stored assessment data from the REM computer storage device to a computer memory controller circuit, and in response to the transmitting, moving the moved computer memory circuit data from the host computer storage device back to the portion of the computer memory circuit.