Flash Memory Controller Reference Count Data Rearrangement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional de-duplication techniques for flash memory storage devices are inefficient due to their reliance on hard disk drive operational methods, which are not suitable for flash memories that cannot overwrite data, leading to challenges in managing data storage and erasure in mobile devices with increasing data storage needs.

Innovation Solution

A storage device with first and second memory blocks and a controller that rearranges data based on reference counts, using a flash translation layer to map logical addresses to physical addresses, allowing for efficient de-duplication and garbage collection operations, thereby optimizing data storage and erasure in flash memory systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional hard disk drive-based de-duplication techniques are used in flash memory storage devices, then data storage management can be simplified, but data transfer efficiency deteriorates and storage capacity utilization is reduced due to flash memory's inability to overwrite data

Engineering Contradiction:
Improvedata storage managementVSAvoiddata transfer efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent changes the parameter of data organization by introducing reference counts to track data sharing levels. Data is categorized into different groups based on reference counts (e.g., unique data with count=1, shared data with count>1), enabling the system to adapt its management strategy to the specific characteristics of flash memory while maintaining efficient de-duplication operations

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent performs preliminary actions by pre-establishing reference count tracking mechanisms and pre-organizing data into distinct groups before erasure operations are needed. This allows the controller to identify which data blocks need to be moved during garbage collection, avoiding unnecessary data transfers and optimizing the erasure process

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If data is frequently rearranged during erasure operations to manage reference counts, then data management accuracy is improved, but storage capacity utilization deteriorates due to increased data transfer overhead

Engineering Contradiction:
Improvedata management accuracyVSAvoidstorage capacity utilization
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies partial action by selectively moving only the necessary data blocks during garbage collection operations. Instead of rearranging all data, the controller identifies specific blocks with changed reference counts and moves only those, reducing unnecessary data transfer while maintaining accurate reference count management

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The system performs self-service by automatically tracking reference counts and managing its own data organization. The controller monitors reference count changes and autonomously determines when data needs to be moved during garbage collection, eliminating the need for external intervention and optimizing storage capacity utilization

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS8898414B2Storage devices and methods of driving storage devices
Publication Date: 2014.11.25 SAMSUNG ELECTRONICS CO LTD
  • US8898414B2 patent drawing
  • US8898414B2 patent drawing
  • US8898414B2 patent drawing

AI summary

A storage device includes a data storage having first and second storage areas corresponding to different physical addresses. First data are stored in the first storage area. The storage device further includes a first memory that stores a reference count associated with the first data, and a controller that rearranges the first data from the first storage area to the second storage area in response to a change in the reference count of the first data.