Distributed File Location Management in Storage Systems

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Solid-state drives designed to conform to hard disk drive standards fail to leverage the unique characteristics of flash and other solid-state memories, limiting their ability to provide enhanced features and efficient data distribution.

Innovation Solution

A storage system architecture that distributes file location information across multiple storage nodes using a hash calculation and lookup table, allowing each storage node to determine the owner of a file based on its name and inode number, thereby avoiding bottlenecks and enhancing efficiency in large directory management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If file location information is centralized in a single storage node, then directory management is simplified, but system reliability deteriorates due to single point of failure and performance bottlenecks

Engineering Contradiction:
Improvedirectory management complexityVSAvoidsystem reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the centralized directory information into distributed chunks across multiple storage nodes. Each storage node maintains a portion of the file location information, eliminating the single point of failure while preserving directory management functionality through distributed coordination

Inventive Principle:
Principle #1Segmentation

2Ease of operation

If all directory information is stored in one storage node, then data access is simplified, but system performance deteriorates due to bottlenecks

Engineering Contradiction:
Improvedata access simplicityVSAvoidsystem performance
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

Directory information is segmented and distributed across multiple storage nodes, allowing parallel access to different file location data simultaneously. This eliminates the bottleneck at a single node while maintaining simplified access through standardized lookup procedures

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-dimension centralized storage model to a multi-dimensional distributed model, where file location information can be accessed from multiple storage nodes concurrently, significantly improving system throughput and performance

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If file location information is distributed across multiple storage nodes, then system reliability is improved, but device complexity increases

Engineering Contradiction:
Improvesystem reliabilityVSAvoidinformation distribution complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements feedback mechanisms where storage nodes communicate file location information to each other, enabling automatic updates and consistency maintenance across the distributed system. This reduces the perceived complexity by allowing the system to self-manage the distributed information

Inventive Principle:
Principle #23Feedback

4Ease of operation

If a centralized authority manages all files, then ownership determination is simplified, but loss of time increases due to single-point processing

Engineering Contradiction:
Improveownership determination simplicityVSAvoidfile access time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

Ownership determination authority is segmented and distributed across multiple storage nodes. Each node can independently determine ownership for files within its jurisdiction, enabling parallel ownership checks and significantly reducing access time while maintaining simplified determination logic through standardized rules

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20220107920A1File location for distributed directory
Publication Date: 2022.04.07 PURE STORAGE INC
  • US20220107920A1 patent drawing
  • US20220107920A1 patent drawing
  • US20220107920A1 patent drawing

AI summary

A method for distribution of directories in a storage system is provided. The method includes distributing information, regarding location in the storage system of a plurality of files in a directory, to a plurality of owners in a plurality of storage nodes of the storage system, wherein ownership of differing subsets of the plurality of files of the directory is distributed among differing owners in differing storage nodes. The method includes receiving a request for location information in the storage system of a file of the plurality of files in the directory and determining, based on a file name of the file and an identity of the directory, which of the plurality of owners has ownership of the file and the location information for the file.