Memory Subsystem Recovery Management for Retired Super Management Units

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing memory sub-systems face inefficiencies in managing retired super management units (SMUs), where entire SMUs are retired due to a single defective management unit, leading to excessive loss of usable media and reduced user capacity, as non-defective MUs are also retired along with defective ones.

Innovation Solution

The memory sub-system implements a recovery management component that identifies defective MUs within retired SMUs, applies appropriate program pulses to restore non-defective MUs to operational status, and removes SMUs from the retired list if all MUs are determined to be non-defective, thereby minimizing unnecessary retirement.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If entire SMUs are retired due to a single defective MU, then reliability is improved by removing defective units, but usable media capacity is excessively reduced by retiring non-defective MUs along with defective ones

Engineering Contradiction:
ImprovereliabilityVSAvoidusable media capacity
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent segments the SMU retirement process by identifying and isolating defective MUs from non-defective MUs within the same SMU. Instead of retiring the entire SMU, the system selectively retires only the defective MUs while preserving non-defective ones, thereby resolving the contradiction between maintaining reliability and preserving usable capacity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts defective MUs from the SMU structure and places them in a separate retired list, while leaving the non-defective MUs operational. This extraction approach allows the system to remove only the harmful defective components without unnecessarily retiring functional MUs, thus maintaining both reliability and capacity.

Inventive Principle:
Principle #2Taking out (Extraction)

2Quantity of substance

If program pulses are applied to recover non-defective MUs, then usable media capacity is improved by reactivating recoverable units, but time and processing resources are consumed by the recovery process

Engineering Contradiction:
Improveusable media capacityVSAvoidrecovery processing time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent applies program pulses to MUs in the retired list before final retirement is executed, as a preliminary recovery attempt. This preliminary action allows the system to recover potentially fixable MUs early in the process, increasing usable capacity while limiting the time investment to only those MUs that might be recoverable.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the operational state of MUs by applying program pulses that modify electrical parameters (voltage, current) to attempt recovery. By adjusting these parameters temporarily to test for recoverability, the system can identify and reactivate MUs that were incorrectly retired, thereby increasing capacity without permanently committing excessive time to unrecoverable units.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11929138B2Recovery management of retired super management units
Publication Date: 2024.03.12 MICRON TECHNOLOGY INC
  • US11929138B2 patent drawing
  • US11929138B2 patent drawing
  • US11929138B2 patent drawing

AI summary

A system includes a memory component, and a processing device coupled with the memory component. The processing device to identify a group of management units of the memory component, wherein the group of management units is included in a set of retired groups of management units, select a management unit from the group of management units, perform a media integrity check on the management unit to determine a failed bit count of the management unit, and in response to the failed bit count of the management unit failing to satisfy a threshold criterion, remove the group of management units from the set of retired groups of management units.