Distributed Storage Node Selection via Probability Factors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing distributed data storage systems face challenges in achieving robust and reliable data writing while maintaining system integrity, particularly during intensive operations and node malfunctions.

Innovation Solution

A method involving a server that sends a multicast storage query to multiple storage nodes, selects nodes based on an algorithmic probability factor, and randomly selects storage nodes for data writing, ensuring geographic diversity and reducing the risk of overloading high-ranked nodes, thereby facilitating quicker replication and maintenance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If storage nodes are selected based on high rank for data writing, then writing reliability is improved, but storage nodes become overloaded during intensive operations

Engineering Contradiction:
Improvewriting reliabilityVSAvoidstorage node capacity
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent changes the selection parameter from static rank-based selection to dynamic probability-based selection. Storage nodes are selected based on a probability factor that considers both their rank and current load status, allowing the system to adapt to changing conditions and avoid overloading high-ranked nodes during intensive operations

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The selection process transitions from a static ranking system to a dynamic probabilistic system. The probability factor is recalculated based on current system state, enabling flexible adaptation to varying load conditions while maintaining reliability through intelligent node selection

Inventive Principle:
Principle #15Dynamics

2Device complexity

If data is concentrated on fewer storage nodes, then maintenance complexity is reduced, but replication time increases when nodes malfunction

Engineering Contradiction:
Improvemaintenance complexityVSAvoidreplication time
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The system proactively distributes data to multiple storage nodes before failures occur. By using probability-based selection to spread data across multiple nodes in advance, the system ensures that replication can be quickly performed from multiple sources if any node malfunctions, rather than concentrating data on fewer nodes

Inventive Principle:
Principle #10Preliminary action

3Reliability

If storage nodes are geographically distributed, then system redundancy is improved, but data spreading efficiency decreases

Engineering Contradiction:
Improvesystem redundancyVSAvoiddata spreading efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent introduces a probability factor that combines geographic distribution requirements with efficiency considerations. Instead of purely random or strictly geographic selection, the system calculates probability factors that balance both redundancy and efficiency, allowing flexible adaptation to different operational requirements

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2712149B1Distributed data storage
Publication Date: 2019.10.30 COMPUVERDE
  • EP2712149B1 patent drawingFigure 1
  • EP2712149B1 patent drawingFigure 2A~2C
  • EP2712149B1 patent drawingFigure 3

AI summary

The present invention relates to a distributed data storage system comprising a plurality of storage nodes. Using unicast and multicast transmission, a server application may write data in the storage system. When writing data, at least two storage nodes are selected based in part on a randomized function, which ensures that data is sufficiently spread to provide efficient and reliable replication of data in case a storage node malfunctions.