Standby Storage Units for Dispersed Network Failure Recovery

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current dispersed storage networks face challenges in maintaining data integrity and availability due to storage unit failures, as they require redundant copies and are prone to data loss when multiple units fail, especially in large-scale systems where timely replacement is necessary.

Innovation Solution

The implementation of a dispersed storage network with error encoding using Cauchy Reed-Solomon encoding and a managing unit that coordinates data distribution across multiple storage units, allowing for data reconstruction from a subset of encoded slices and proactive management of standby units to replace failing ones, ensuring data availability and security.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If redundant copies are used for data storage, then data availability is improved, but storage space efficiency deteriorates

Engineering Contradiction:
Improvedata availabilityVSAvoidstorage space efficiency
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The data is divided into multiple slices and distributed across different storage units. Instead of storing complete redundant copies, the system segments data into portions that can be recombined to reconstruct the original data, improving storage efficiency while maintaining availability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs error encoding schemes that transform data into encoded slices with specific parameters. By changing the representation parameters of data (from direct copies to encoded segments), the system achieves both reliability and space efficiency through mathematical reconstruction capabilities.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If multiple storage units are used for data distribution, then system scalability is improved, but data loss risk worsens when multiple units fail

Engineering Contradiction:
Improvesystem scalabilityVSAvoiddata loss risk
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system proactively identifies and replaces storage units before failures occur. By monitoring storage unit health and preemptively replacing units showing signs of failure, the system prevents data loss scenarios while maintaining scalability across multiple distributed units.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements monitoring and feedback mechanisms that track the status of distributed storage units. This feedback system enables the network to detect potential failures and trigger replacement actions, ensuring data integrity even as the system scales to include more storage units.

Inventive Principle:
Principle #23Feedback

3Reliability

If standby storage units are maintained for replacement, then system reliability is improved, but device complexity worsens

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

Solution Approach 1:

Standby storage units are designed with multi-functionality, serving both as replacement units for failed storage and as active storage resources when needed. This universal approach reduces the need for dedicated spare units, lowering overall system complexity while maintaining reliability.

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

Solution Approach 2:

The system implements automated processes for identifying failed storage units, selecting appropriate standby replacements, and redistributing data slices. This self-service capability reduces manual intervention requirements and simplifies the management complexity of maintaining standby units for system reliability.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS10423502B2Stand-by distributed storage units
Publication Date: 2019.09.24 PURE STORAGE INC
  • US10423502B2 patent drawing
  • US10423502B2 patent drawing
  • US10423502B2 patent drawing

AI summary

A method for execution by a dispersed storage network (DSN), the method begins by determining a failure rate of storage units of an active storage unit pool, establishing a number of standby storage units based on the determined failure rate, identifying an associated DSN address range of the failed storage unit, selecting an available standby storage unit, facilitating populating the selected available standby storage unit with data slices associated with the failed storage unit, utilizing the selected available standby storage unit, facilitating population of a replacement storage unit with the data slices from the selected available standby storage unit and facilitating processing of further receive data access requests for data associated with the associated DSN address range by utilizing the replacement storage unit.