DS Processing Unit Priority Ordering for Dispersed Storage Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current distributed storage and task processing systems face challenges in securely and reliably storing and retrieving large amounts of data across geographically dispersed locations, while ensuring data integrity and tolerating failures without the need for redundant copies.
Innovation Solution
A distributed computing system that employs dispersed error encoding and decoding, where data is segmented, encoded, and distributed across multiple storage units, allowing for secure, reliable storage and retrieval, and tolerant of failures through error correction mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is distributed across multiple storage units using error correction encoding, then reliability and fault tolerance are improved, but device complexity increases due to encoding and decoding operations
Solution Approach 1:
The patent segments data into multiple slices and distributes them across different storage units. Each slice is encoded using error correction codes, allowing the system to reconstruct original data even if some storage units fail. This segmentation approach improves reliability by eliminating single points of failure while managing complexity through modular encoding operations.
Solution Approach 2:
The patent introduces encoding functions as intermediaries between the data and storage units. These encoding functions transform original data into encoded slices that can be distributed and later decoded to recover the original information. The intermediary encoding layer protects data integrity without requiring redundant copies, resolving the contradiction between reliability and storage efficiency.
2Reliability
If data is segmented and distributed across geographically dispersed storage units, then availability and fault tolerance are improved, but loss of time increases due to coordinated access across multiple locations
Solution Approach 1:
The patent divides data into multiple slices that can be independently stored and accessed across geographically dispersed storage units. This segmentation allows parallel access operations, where multiple slices can be retrieved simultaneously from different locations, reducing the overall access time compared to sequential access of centralized data.
Solution Approach 2:
The patent implements error correction encoding that allows data reconstruction from a subset of slices rather than requiring all slices. This partial action principle enables the system to retrieve data from fewer storage units than the total number of distributed slices, significantly reducing access time while maintaining data availability and integrity.
3Quantity of substance
If error correction encoding is used to tolerate failures without redundant copies, then storage efficiency is improved, but device complexity increases due to encoding and decoding operations
Solution Approach 1:
The patent extracts the error correction functionality into separate encoding and decoding operations that are applied to data slices. This extraction allows the use of efficient error correction codes that provide strong fault tolerance with minimal overhead, improving storage efficiency by eliminating the need for full redundant copies while managing complexity through specialized processing functions.
Solution Approach 2:
The patent employs error correction encoding schemes that change the parameters of data representation (e.g., adding checksum bits, using polynomial codes) to enable failure tolerance. These parameter changes are applied through standardized encoding functions that achieve high storage efficiency with controlled processing complexity, as the encoding overhead is minimal compared to traditional redundancy approaches.
Data Source
AI summary
A method includes receiving an DSN access request, identifying an DSN address of the DSN access request, identifying one or more DS processing units affiliated with the DSN address. The method further includes selecting a DS processing unit of the one or more DS processing units based on DS processing attributes, determining if the selected DS processing unit is associated with a favorable availability level and, when it is determined that a favorable availability level does not exist, deterministically selecting another of the one or more DS processing units and, when it is determined that a favorable availability level does exist, forwarding the DSN access request to the selected DS processing unit to be processed for storage in one or more DSN storage units.