Controller Access Count Hashing for Nonvolatile Memory Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data storage devices face challenges in maintaining data reliability due to excessive access leading to deformation or loss in neighboring regions, as existing technologies lack effective mechanisms for real-time monitoring and maintenance of access counts across nonvolatile memory apparatuses.

Innovation Solution

A data storage device with a controller that updates access counts and performs reliability maintenance operations by using a hashing unit to map addresses into hash values, managing access counts in an integrated manner, and performing data inspection and reclaim operations based on access thresholds to prevent data deformation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If access count monitoring is implemented for each region, then data reliability is improved, but device complexity increases

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

Solution Approach 1:

The patent combines access count management for multiple regions into a unified system using hash value mapping. Instead of implementing separate monitoring mechanisms for each region, the controller uses a hashing function to map multiple region addresses to shared access count tables, merging the monitoring function into an integrated structure that reduces overall system complexity while maintaining comprehensive reliability monitoring.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces hash values as an intermediary layer between physical memory regions and access count records. This intermediary abstraction allows the system to monitor access counts without direct one-to-one mapping between regions and counters, simplifying the monitoring infrastructure while enabling effective data reliability management through indirect but systematic tracking.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If real-time access count monitoring is performed, then data deformation is prevented, but processing time increases

Engineering Contradiction:
Improveprevention of data deformationVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent performs preliminary hashing of region addresses to generate hash values that can be used for access count lookup. By pre-computing and storing the mapping between region addresses and hash values, the system avoids time-consuming address calculations during actual access operations, enabling real-time monitoring with minimal processing overhead.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements periodic initialization of access counts based on hash value patterns. Instead of continuously updating all access counts for every single access, the system periodically resets counts for regions with identical hash values, reducing the frequency of update operations while maintaining effective monitoring and prevention of data deformation.

Inventive Principle:
Principle #19Periodic action

3Measurement precision

If multiple access count tables are maintained, then monitoring precision is improved, but memory usage increases

Engineering Contradiction:
Improvemonitoring precisionVSAvoidmemory usage
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the memory space into regions that share common hash values, allowing access count tables to be organized by hash value groups rather than by individual physical regions. This segmentation strategy enables precise tracking of access patterns for each hash group while reducing the total number of access count entries needed, as multiple regions map to the same count table entries.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different access count management strategies to different hash value groups. Regions with identical hash values share the same access count table entries, while regions with different hash values have separate entries. This local quality approach optimizes memory usage by allocating monitoring resources according to the actual diversity of access patterns rather than uniformly across all regions.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS9678827B2Access counts for performing data inspection operations in data storage device
Publication Date: 2017.06.13 SK HYNIX INC
  • US9678827B2 patent drawing
  • US9678827B2 patent drawing
  • US9678827B2 patent drawing

AI summary

A data storage device includes a controller configured to update an access request count and an access count corresponding to a target region based on an access request for the target region, and initialize the access count each time the access request count reaches a first threshold, and a nonvolatile memory apparatus including the target region, and configured to access the target region based on a control of the controller.