Dispersed Storage Network Write Threshold Adaptation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing distributed storage and task processing systems face challenges in efficiently managing and processing large datasets across multiple geographically dispersed locations, particularly in ensuring data consistency and integrity while supporting distributed task execution.

Innovation Solution

The system employs a distributed computing architecture that includes a user device, a distributed storage and task (DST) processing unit, a DST network managing unit, and a DST integrity processing unit. It uses dispersed error encoding to store data and execute tasks across multiple DST execution units, ensuring data integrity and consistency through error correction and secure encoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is stored across multiple geographically dispersed locations in a distributed storage system, then system availability and fault tolerance are improved, but data consistency and integrity become more difficult to ensure

Engineering Contradiction:

Improvesystem availabilityVSAvoiddata consistency

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent segments data into multiple slices and distributes them across different storage units in the dispersed storage network. Each slice is stored independently, allowing the system to maintain availability even when some storage units fail, while consistency is maintained through the encoding relationship between slices

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements feedback mechanisms where storage units send status information and data integrity checks back to the controlling entity. This allows the system to detect and correct consistency issues, ensuring data integrity across distributed locations while maintaining high availability

Inventive Principle:

Principle #23Feedback

2Productivity

If distributed task processing is implemented across multiple locations, then processing capacity and scalability are improved, but coordination overhead and system complexity increase

Engineering Contradiction:

Improveprocessing capacityVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent creates a universal dispersed storage interface that can handle multiple task types and data operations through a common framework. This multi-functional approach allows distributed task processing across multiple locations without requiring separate coordination mechanisms for each task type, thereby scaling productivity while controlling system complexity

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces intermediary components including the dispersed storage interface and controlling entity that mediate between distributed storage units and processing tasks. These intermediaries abstract the complexity of distributed coordination, allowing tasks to be processed across multiple locations with reduced overhead

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If error correction encoding is applied to ensure data integrity, then data security and fault tolerance are improved, but storage overhead and processing time increase

Engineering Contradiction:

Improvedata integrityVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies error correction encoding at selective points in the data flow rather than continuously. The dispersal encoding is applied when data is written to the distributed storage system, and verification is performed only when needed during task processing or consistency checks, reducing overall processing time while maintaining data integrity

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The patent performs error correction encoding in advance when data is initially stored in the dispersed storage network. This preliminary encoding ensures data integrity is built-in from the start, eliminating the need for repeated encoding operations during subsequent processing tasks, thereby reducing processing time while maintaining high data integrity

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250138747A1Write Request Fulfillment in a Storage Network

Publication Date: 2025.05.01 PURE STORAGE INC

AI summary

A processing system of a storage network operates by receiving a write request associated with a data object; identifying dispersed storage error encoding parameters; and determining a number of available storage units. When the number of available storage units is determined to exceed a first number, the processing system operates by: determining a first write threshold number in accordance with the dispersed storage error encoding parameters; dispersed error encoding the data object to produce a first write threshold number of encoded data slices; and writing, to available storage units, the first write threshold number of encoded data slices. When the number of available storage units is determined to not exceed the first number, the processing system operates by: determining a second write threshold number in accordance with the dispersed storage error encoding parameters associated, wherein the second write threshold number is less than the first write threshold number; dispersed error encoding the data object to produce a second write threshold number of encoded data slices; and writing, to the available storage units, the second write threshold number of encoded data slices.