Memory Sub-System Controller Partial Block Failure Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current memory sub-systems often treat a partial block failure as a complete block failure, leading to unnecessary data relocation and garbage collection operations, which can impact performance and reliability by causing write amplification and reducing Quality of Service (QOS).

Innovation Solution

Implementing a memory sub-system controller that intelligently identifies and prioritizes data from partially failed blocks, using known-failure modes to determine the severity of failures and selectively fold only the failed portions of a block, allowing non-failed parts to remain operational until garbage collection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a partial block failure is treated as a complete block failure, then data safety is improved, but write amplification increases and performance deteriorates

Engineering Contradiction:
Improvedata safetyVSAvoidperformance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the block into failed and non-failed portions, allowing selective handling of only the failed segments rather than treating the entire block as failed. This enables continued operation of healthy memory cells while isolating defective areas.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different quality levels to different parts of the block - failed portions receive error handling and data relocation, while non-failed portions continue normal operation without unnecessary interference, optimizing overall system performance.

Inventive Principle:
Principle #3Local quality

2Reliability

If a partial block failure is treated as a complete block failure, then data safety is improved, but unnecessary data relocation operations increase

Engineering Contradiction:
Improvedata safetyVSAvoidwrite amplification
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The patent extracts only the failed portion from the block for error handling, separating it from the functional non-failed portion. This extraction enables targeted data relocation only where necessary, avoiding redundant operations on healthy data.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by performing error handling operations only on the failed subset of the block rather than the entire block, reducing unnecessary write amplification while maintaining data safety for affected areas.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If a partial block failure is treated as a complete block failure, then data safety is improved, but garbage collection operations increase

Engineering Contradiction:
Improvedata safetyVSAvoidQuality of Service
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the block to identify and isolate failed portions, allowing the garbage collection process to skip over healthy non-failed portions and focus only on areas requiring data relocation, thereby reducing overall garbage collection time.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial garbage collection operations only to the failed portion of the block rather than the entire block, maintaining data safety for affected areas while minimizing the time impact on Quality of Service.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20240355402A1Known-failure error handling in a memory sub-system
Publication Date: 2024.10.24 MICRON TECHNOLOGY INC
  • US20240355402A1 patent drawing
  • US20240355402A1 patent drawing
  • US20240355402A1 patent drawing

AI summary

A system and method for a memory device for detecting, by a processing device, a failure exhibited by a set of cells of a memory device, estimating a severity of the failure, identifying, based on the severity of the failure, a failed subset of cells of the set of cells, and copying data from the failed subset of cells to a second set of cells of the memory device.