NAND Volume Partitioning for Mixed-Reliability Data Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current non-volatile memory devices face challenges in efficiently managing both high and normal reliability data on a single device, with existing multi-mode MLC memory systems prone to errors and requiring external file management systems, which complicates data storage and updates.

Innovation Solution

A non-volatile memory device with a volume management system that dynamically partitions MLC and SLC regions, utilizing a meta-volume directory for centralized data management, error correction codes, and wear-leveling to ensure data integrity and adaptability, allowing for simultaneous storage of high and normal reliability data without external management systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If MLC NAND architecture is used to store both user data and high reliability data, then storage capacity is increased and device size is reduced, but data reliability deteriorates

Engineering Contradiction:
Improvestorage capacityVSAvoiddata reliability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The memory device is segmented into distinct functional regions: a first region configured to store high reliability data (such as OS data and boot code) and a second region configured to store user data. This segmentation allows each region to be optimized for its specific purpose, with the first region providing enhanced reliability through dedicated error correction and verification mechanisms while the second region maximizes storage capacity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different quality levels are applied to different parts of the memory system. The first region implementing a localized file management system with enhanced error correction, verification protocols, and data integrity checks provides higher reliability quality, while the second region uses standard MLC storage providing adequate quality for user data. This local quality differentiation resolves the contradiction by providing high reliability where needed without compromising overall storage capacity.

Inventive Principle:
Principle #3Local quality

2Reliability

If conventional approach stores OS data in highly reliable location separate from user data, then data reliability is improved, but device complexity and size increase

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

Solution Approach 1:

The patent merges the file management system into the memory device itself rather than using external management systems. The localized file management system is integrated within the memory device, combining high reliability data storage and user data storage into a single unified device. This integration reduces overall system complexity by eliminating external management requirements while maintaining data reliability through the specialized first region configuration.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The memory device is designed with multi-functionality, serving both as high reliability storage for OS data and as general user data storage. The single memory device performs multiple functions: storing boot code, OS data, and user data, while providing its own integrated file management capabilities. This universality reduces device complexity by consolidating multiple functions into one device rather than requiring separate dedicated storage devices.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Quantity of substance

If MLC memory stores multiple bits per cell, then storage density is improved, but error rate increases

Engineering Contradiction:
Improvestorage densityVSAvoiderror rate
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The first region is configured with beforehand cushioning through enhanced error correction mechanisms and verification protocols. Data integrity checks and error correction codes are implemented in advance before data is considered reliably stored. This prior cushioning compensates for the inherently higher error rate of MLC memory by providing built-in protection mechanisms that detect and correct errors before they affect data usability.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Solution Approach 2:

The patent applies parameter changes by implementing different storage parameters for the first region versus the second region. The first region uses more conservative programming parameters, lower density configurations, and enhanced verification thresholds to reduce error rates. The second region uses aggressive MLC parameters for maximum density. This parameter differentiation allows the system to achieve high storage density in the second region while maintaining low error rates in the first region where reliability is critical.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10083078B2Method and apparatus for a volume management system in a non-volatile memory device
Publication Date: 2018.09.25 MICRON TECHNOLOGY INC
  • US10083078B2 patent drawing
  • US10083078B2 patent drawing
  • US10083078B2 patent drawing

AI summary

Embodiments for partitioning a non-volatile memory device is described. In one embodiment a memory system includes a first addressable range of memory blocks for storing different types of data. The memory system is partitioned to include a second addressable range of memory blocks capable of storing data indicating attributes of the first addressable range of memory blocks. The second addressable range of memory blocks may also be periodically updated such that the capacities of the first addressable range of memory blocks may be dynamically adjusted depending on application needs and changes to the non-volatile memory device over time In some embodiments, one partition of a memory device may be configured for high reliability data storage while a second partition is configured for normal reliability storage.