Storage Control for Data De-Duplication With Regenerated Check Codes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data de-duplication techniques in storage systems face challenges in maintaining reliability when check codes are separated from data, leading to decreased error detection capabilities in compressed data, and existing methods do not adequately consider de-duplication during data changes.

Innovation Solution

A storage control apparatus that separates first-code attached data from error detection codes and positional information, eliminates data duplication, generates new error detection codes for de-duplicated data, and writes these codes back into the memory device, ensuring data integrity and capacity efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If check code is separated from data to enable de-duplication, then capacity efficiency is improved, but error detection capability deteriorates

Engineering Contradiction:
Improvedata capacity efficiencyVSAvoiderror detection capability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent segments the check code generation process into two parts: first generating a check code from the original data before de-duplication, then generating a dummy check code from the de-duplicated data. This segmentation allows the system to maintain error detection capability while achieving de-duplication benefits.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of check code generation by using different input data (original data vs. de-duplicated data) and adjusting the check code accordingly. This parameter change enables the system to adapt to the de-duplicated state while preserving reliability through dummy code generation.

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If de-duplication is performed on data with check code, then data reduction is achieved, but the check code becomes invalid for compressed data

Engineering Contradiction:
Improvedata amount reductionVSAvoidcheck code validity
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent performs preliminary check code generation from the original data before de-duplication occurs. This preliminary action ensures that a valid check code exists for the original data, and a corresponding dummy check code is generated for the de-duplicated data, maintaining check code validity throughout the process.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a dummy check code that copies the structure and function of the original check code but is generated from the de-duplicated data. This copying approach ensures that the check code format and validation mechanism remain consistent even after de-duplication transforms the data.

Inventive Principle:
Principle #26Copying

3Ease of manufacture

If existing check code methods are used without considering de-duplication, then implementation simplicity is maintained, but de-duplication effectiveness is reduced

Engineering Contradiction:
Improveimplementation simplicityVSAvoidde-duplication effectiveness
Core Design Contradiction:
Ease of manufactureVSQuantity of substance

Solution Approach 1:

The patent creates a universal check code generation mechanism that works for both original data and de-duplicated data. By generating dummy check codes that follow the same validation logic, the system maintains a single, unified approach to error detection that applies regardless of whether the data has been de-duplicated, thus preserving implementation simplicity while enhancing de-duplication effectiveness.

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

Data Source

PatentEP3106986B1Storage control apparatus and program
Publication Date: 2018.01.03 FUJITSU LTD
  • EP3106986B1 patent drawingFigure 1
  • EP3106986B1 patent drawingFigure 2
  • EP3106986B1 patent drawingFigure 3

AI summary

There is provided a storage control apparatus (10) that obtains first-code attached data (CD00, ..., CD23), each having target data (D0, ···, D2) to be written and first code information (C0, ···, C3), which includes an error detection code based on the target data (D0, ..., D2) and information about a first write destination, attached to the target data (D0, ···, D2); obtaining the target data (D0, ..., D2) by excluding the first code information (C0, ..., C3) from the first-code attached data (CD00, ···, CD23); eliminating duplication of the target data (D0, ..., D2); generating second code information (C4, C5, C6) which includes an error detection code for the target data (D0, D1, D2) remaining and information about a second write destination; and writing second-code attached data (CD04, CD15, CD26) including the second code information (C4, C5, C6) into a memory device (20).