Dispersed Storage Network Error Recovery Data Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current distributed storage networks face challenges in ensuring data integrity and security, particularly in maintaining error correction and secure signature generation across multiple storage units, which can lead to data loss and unauthorized access.

Innovation Solution

A dispersed storage network (DSN) architecture that employs error encoding using Cauchy Reed-Solomon encoding, distributing data into multiple encoded slices stored across geographically diverse units, with an integrity processing unit for rebuilding corrupted slices and a managing unit for secure key share generation and signature creation, ensuring data integrity and security.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is distributed across multiple storage units for redundancy and security, then data reliability and security are improved, but system complexity increases

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

Solution Approach 1:

The patent segments data into multiple encoded slices using Cauchy Reed-Solomon encoding, distributing them across different storage units. Each slice is an independent part of the whole, allowing the system to tolerate failures of individual storage units while maintaining data integrity. The segmentation principle directly resolves the contradiction by improving reliability through distribution while managing complexity through standardized encoding procedures.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an integrity processing unit as an intermediary component that automatically detects and repairs corrupted slices. This mediator handles the complexity of error correction and data recovery, shielding the user from system complexity while maintaining high data integrity through automated repair processes.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If error correction encoding is applied to protect data, then data security is improved, but processing time increases

Engineering Contradiction:
Improvedata securityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies Cauchy Reed-Solomon encoding in advance during the data writing process, so that error correction capability is built into the stored slices before any potential corruption occurs. This preliminary encoding action ensures data security is established upfront, avoiding time-consuming error correction during read operations and thus reducing processing time loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The encoded slices contain redundant information that enables self-correction of errors without requiring complex external intervention. The integrity processing unit uses the encoded structure to automatically detect and repair corrupted slices, making the error correction process efficient and minimizing additional processing time while maintaining data security.

Inventive Principle:
Principle #25Self-service

3Reliability

If multiple key shares are stored across different units for secure signature generation, then security against unauthorized access is improved, but key management complexity increases

Engineering Contradiction:
ImprovesecurityVSAvoidkey management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the private key into multiple key shares using secret sharing schemes, storing them across different storage units. No single unit holds the complete key, providing security through distribution. The segmentation principle improves security while managing complexity by using standardized key sharing algorithms that automate the distribution and reconstruction processes.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a universal key management system where the same encoding and decryption mechanisms handle both data storage and signature generation operations. The managing unit can reconstruct key shares from any sufficient subset of storage units, providing a multi-functional solution that handles security operations uniformly, thus reducing key management complexity through standardization.

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

4Reliability

If data is encoded and distributed across geographically diverse units, then fault tolerance is improved, but network communication overhead increases

Engineering Contradiction:
Improvefault toleranceVSAvoidnetwork communication overhead
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent uses Cauchy Reed-Solomon encoding to create more encoded slices than the minimum required for reconstruction. This excessive encoding provides enhanced fault tolerance by allowing recovery from a greater number of failures, while the encoding efficiency minimizes the additional network overhead. The system retrieves only the necessary number of slices for reconstruction, optimizing network communication.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The integrity processing unit implements feedback mechanisms to monitor the health of distributed slices and proactively initiate repair operations when corruption is detected. This feedback system optimizes network communication by triggering repairs only when necessary, rather than continuously synchronizing all units, thus reducing network overhead while maintaining fault tolerance across geographically diverse locations.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS20220368757A1Managing Error Recovery Data in a Dispersed Storage Network
Publication Date: 2022.11.17 PURE STORAGE INC
  • US20220368757A1 patent drawing
  • US20220368757A1 patent drawing
  • US20220368757A1 patent drawing

AI summary

A method for managing error recovery data in a dispersed storage network begins with a storage network processing module receiving a write request for an encoded data slice of a set of encoded data slices, where data is dispersed in accordance with dispersed error encoding parameters to produce a set of encoded data slices. The method continues with the storage network processing module generating parity data for the encoded data slice and sending the encoded data slice to a first storage unit of a set of storage units. Finally, the method continues with the storage network processing module sending the parity data for the encoded data slice to a second storage unit of a set of storage units.