Distributed Storage Data Migration for Fault Tolerance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In distributed storage systems, device failures, such as those caused by high temperatures, can prevent access to stored data, either temporarily or permanently, leading to data unavailability and accessibility issues.

Innovation Solution

A method that detects anomalies on devices and transfers data to alternative storage devices within the system, updating the storage location to ensure data availability and accessibility, with options for repatriation and data synchronization upon anomaly resolution or modification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is stored on a single device in a distributed storage system, then storage capacity and accessibility are improved, but data availability deteriorates when device failure occurs

Engineering Contradiction:
Improvedata accessibilityVSAvoiddata availability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system proactively detects anomalies on storage devices and triggers data transfer to alternative devices before complete failure occurs. The management module monitors device health and initiates preventive data migration, ensuring data remains accessible even if the original device fails.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention creates copies of data on alternative storage devices within the distributed system. When an anomaly is detected on a primary device, the management module transfers data to secondary devices, maintaining multiple copies to ensure data availability and accessibility.

Inventive Principle:
Principle #26Copying

2Reliability

If data is transferred to alternative storage devices upon anomaly detection, then data availability is improved, but system complexity increases due to location management

Engineering Contradiction:
Improvedata availabilityVSAvoidstorage location management
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The management module serves as an intermediary that abstracts the complexity of data location management. It maintains knowledge of data locations across devices and handles transfer operations, shielding users and applications from the underlying complexity while ensuring data availability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements feedback mechanisms where the management module continuously monitors device status and adjusts data locations based on detected anomalies. This feedback loop enables automatic data migration to maintain availability without requiring complex manual management.

Inventive Principle:
Principle #23Feedback

3Ease of operation

If data is continuously synchronized across multiple devices, then data accessibility is improved, but energy consumption and system overhead increase

Engineering Contradiction:
Improvedata accessibilityVSAvoidenergy consumption
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

Instead of continuous synchronization, the system uses periodic anomaly detection and event-triggered data transfers. The management module monitors devices at intervals and initiates transfers only when anomalies are detected, reducing energy consumption while maintaining data accessibility when needed.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system employs self-service mechanisms where the management module automatically detects anomalies and triggers data transfers without requiring continuous external intervention. This on-demand approach reduces system overhead and energy consumption compared to continuous active synchronization.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2531921B1Management of location of stored data in a distributed storage system
Publication Date: 2015.05.13 ORANGE SA
  • EP2531921B1 patent drawingFigure 1~2
  • EP2531921B1 patent drawingFigure 3~4

AI summary

The invention relates to a method for managing a content-storage site in a distributed storage system including a plurality of data-processing devices comprising at least one respective processor and a respective storage means in which content has a respective storage site to be used when accessing the content, characterised in that a detection of a fault linked to the use of the device, referred to as the first device, is followed by a step of transferring at least one piece of stored content on said first device towards at least one second device of the distributed storage system in order to be stored in the latter, and in that said method also includes a step of updating the new storage site.