Directory-Less Storage Read Routing With Encoded Data Slices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data storage systems face issues with memory device failures, particularly in commercial-grade devices using physical movement technologies, leading to data loss and security risks, and redundant array of independent discs (RAID) solutions incur efficiency and maintenance challenges as well as security vulnerabilities.

Innovation Solution

A distributed storage network (DSN) system that utilizes error-coded data slices stored across geographically diverse units, ensuring data integrity and security through forward error correction and dispersed storage, managed by a DS managing unit and processed by a DS processing unit for retrieval and integrity verification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If RAID solutions are used to protect against memory device failures, then data reliability is improved, but device complexity and maintenance needs increase

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

Solution Approach 1:

The patent divides data into multiple slices and distributes them across different storage units, implementing segmentation to reduce the complexity burden on any single device while maintaining overall system reliability through distributed architecture

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements nested encoding by applying error correction codes within slices and across slices, creating multiple layers of redundancy that protect data reliability without requiring complex external RAID hardware

Inventive Principle:
Principle #7Nested doll (Nesting)

2Reliability

If RAID solutions are used to protect against memory device failures, then data reliability is improved, but maintenance needs increase

Engineering Contradiction:
Improvedata reliabilityVSAvoidmaintenance needs
Core Design Contradiction:
ReliabilityVSEase of repair

Solution Approach 1:

The patent enables self-service repair through autonomous detection and recovery mechanisms where storage units automatically identify failed slices and initiate reconstruction processes without human intervention, significantly reducing maintenance needs while maintaining data reliability

Inventive Principle:
Principle #25Self-service

3Reliability

If data is distributed across multiple storage units, then security is improved, but device complexity increases

Engineering Contradiction:
ImprovesecurityVSAvoiddevice complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments data into encrypted slices distributed across multiple storage units, enhancing security through distribution while managing complexity through automated slice management and tracking systems

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary encoding layer that manages the complexity of distributed storage by automatically handling slice distribution, tracking, and reconstruction across multiple storage units, thereby simplifying the overall system architecture

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS12531800B2Read requests in a directory-less storage network
Publication Date: 2026.01.20 PURE STORAGE INC
  • US12531800B2 patent drawing
  • US12531800B2 patent drawing
  • US12531800B2 patent drawing

AI summary

A method for execution by a storage unit in a directory-less storage network includes receiving a read request regarding a data segment of a data object, where the data segment was error encoded into a set of encoded data slices. The method further includes outputting a read response regarding an encoded data slice of the set of encoded data slices that is stored by the storage unit. The method further includes identifying an other storage unit of the directory-less storage network to which the read request is to be forwarded, where the identifying the other storage unit includes one or more of accessing a table of localized mapping regarding one or more neighboring storage units to identify the other storage unit, and receiving a message from a network management unit. The method further includes forwarding the read request to the other storage unit.