Dual Data Protection Schemes for Non-Volatile Memory

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Non-volatile memory devices face data retention issues due to physical mechanisms like charge loss in NAND flash and drift/crystallization in phase change memory, which can render system critical data non-functional, and existing error correction schemes are overdesigned or resource-intensive.

Innovation Solution

Implementing a dual data protection scheme for non-volatile memory, where system critical data is written using a first scheme with additional redundancy and error correction, and user data using a second scheme with similar resource allocation, utilizing ECC codes like Reed-Solomon or BCH codes, and employing multiple pulse-verification processes to narrow threshold voltage distributions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If additional levels of protection are implemented for system critical data, then data reliability is improved, but device complexity increases

Engineering Contradiction:
Improvedata reliabilityVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides data protection into two distinct schemes: a first data protection scheme for system critical data and a second data protection scheme for user data. This segmentation allows each scheme to be optimized independently, with the first scheme providing enhanced protection (e.g., stronger ECC codes, additional redundancy) for critical firmware and configuration data, while the second scheme provides adequate protection for regular user files, thereby improving overall data reliability without uniformly increasing complexity across all data types.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different levels of protection quality to different data types based on their criticality. System critical data receives higher-quality protection mechanisms (such as more robust error correction codes, multiple copies, or enhanced verification processes) while user data receives standard protection. This local differentiation ensures that resources are concentrated where they are most needed, improving reliability for critical systems without unnecessarily complicating the protection of less critical user data.

Inventive Principle:
Principle #3Local quality

2Reliability

If multiple copies of system critical data are stored, then data reliability is improved, but memory capacity is consumed

Engineering Contradiction:
Improvedata reliabilityVSAvoidmemory capacity
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent implements selective copying of system critical data with different protection levels. Multiple copies or redundant representations of critical data are stored using efficient encoding schemes that maximize reliability while minimizing the additional memory capacity required. The system intelligently determines which critical data requires multiple copies based on their importance and the available memory resources.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent changes the parameters of data storage by implementing different data protection schemes with varying redundancy levels. For system critical data, parameters such as error correction code strength, redundancy ratio, and verification frequency are adjusted to provide enhanced protection. For user data, these parameters are optimized for memory efficiency. This parameter differentiation allows the system to improve reliability for critical data without proportionally increasing memory consumption across all data types.

Inventive Principle:
Principle #35Parameter changes

3Duration of action of stationary object

If system critical data is stored in SLC format, then data retention is improved, but memory capacity is reduced

Engineering Contradiction:
Improvedata retentionVSAvoidmemory capacity
Core Design Contradiction:
Duration of action of stationary objectVSQuantity of substance

Solution Approach 1:

The patent applies different storage format qualities to different data types. System critical data is stored in SLC (Single-Level Cell) format which provides superior data retention characteristics and lower write amplification, while user data is stored in MLC (Multi-Level Cell) or TLC (Triple-Level Cell) formats that offer higher storage density. This local quality differentiation ensures that the enhanced retention properties of SLC are applied only where critical for system operation, rather than uniformly across all stored data, thereby balancing data retention improvement with memory capacity utilization.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2901292B1Techniques associated with protecting system critical data written to non-volatile memory
Publication Date: 2018.11.21 INTEL CORP
  • EP2901292B1 patent drawingFigure 1
  • EP2901292B1 patent drawingFigure 2
  • EP2901292B1 patent drawingFigure 3

AI summary

Examples are disclosed for techniques associated with protecting system critical data written to non-volatile memory. In some examples, system critical data may be written to a non-volatile memory using a first data protection scheme. User data that includes non-system critical data may also be written to the non-volatile memory using a second data protection scheme. For these examples, both data protection schemes may have a same given data format size. Various examples are provided for use of the first data protection scheme that may provide enhanced protection for the system critical data compared to protection provided to user data using the second data protection scheme. Other examples are described and claimed.