Replicated State Machine Node Induction via Distributed Consensus

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional distributed multi-site collaborative computing solutions face limitations in real-time synchronization, scalability, reliability, and fault-tolerance, leading to delays and increased risk in software development and delivery, particularly due to issues with replicated state machines and information repositories.

Innovation Solution

A replicated state machine architecture with a proposal manager, agreement manager, collision/back-off timer, and storage reclaimer is implemented, allowing for coordinated execution of proposals, preventing repeated pre-emptions, and reclaiming persistent storage, while ensuring all repository replicas are in sync in real-time, eliminating the need for a central transaction coordinator.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional distributed multi-site collaborative computing solutions are used, then distributed resource utilization is achieved, but real-time synchronization capability is lost

Engineering Contradiction:
Improvedistributed resource utilizationVSAvoidreal-time synchronization capability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system segments the centralized transaction coordinator functionality into distributed proposal managers at each node. Each node independently manages its own proposals and participates in the agreement protocol, eliminating the single point of failure while maintaining synchronization through distributed consensus.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary actions by pre-establishing the agreement protocol and proposal mechanisms at each node before actual transactions occur. This allows nodes to be pre-configured with the capability to synchronize in real-time, rather than relying on centralized coordination during operation.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If conventional replicated state machine implementations are used, then state replication is achieved, but scalability is adversely impacted due to repeated pre-emption of proposers

Engineering Contradiction:
Improvestate replicationVSAvoidscalability
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The collision/back-off timer implements preliminary action by detecting potential conflicts before they occur and preemptively delaying proposer actions. This prevents repeated pre-emptions and maintains system scalability while ensuring reliable state replication through controlled conflict resolution.

Inventive Principle:
Principle #10Preliminary action

3Reliability

If conventional replicated state machine implementations are used, then state replication is achieved, but complexity increases due to weak leader optimization requiring leader election

Engineering Contradiction:
Improvestate replicationVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system extracts the leader election mechanism from the state replication process. By removing the complex leader election and weak leader optimization requirements, the system achieves state replication through simpler distributed agreement protocols, reducing overall system complexity while maintaining reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

4Productivity

If conventional distributed multi-site collaborative computing solutions are used, then distributed computing capability is achieved, but fault-tolerance is limited

Engineering Contradiction:
Improvedistributed computing capabilityVSAvoidfault-tolerance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system segments the coordination functionality into distributed proposal managers at each node, eliminating the single point of failure inherent in centralized coordinators. Each node can independently propose and participate in agreement, providing fault-tolerance while maintaining distributed computing capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The storage reclaimer implements discarding and recovering by automatically identifying and reclaiming persistent storage that is no longer needed for state replication. This maintains fault-tolerance by ensuring storage availability while supporting distributed computing operations.

Inventive Principle:
Principle #34Discarding and recovering

5Productivity

If conventional distributed multi-site collaborative computing solutions are used, then distributed resource utilization is achieved, but storage needs grow continuously without bound

Engineering Contradiction:
Improvedistributed resource utilizationVSAvoidpersistent storage requirements
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The storage reclaimer implements automatic discarding and recovering of persistent storage by identifying state data that is no longer needed for replication and reclaiming it. This prevents storage from growing without bound while maintaining the distributed resource utilization capability, as the system can dynamically manage storage resources across the distributed network.

Inventive Principle:
Principle #34Discarding and recovering

Data Source

PatentEP2939447B1Induction of a node into a group
Publication Date: 2019.12.04 WANDISCO
  • EP2939447B1 patent drawingFigure 1
  • EP2939447B1 patent drawingFigure 2
  • EP2939447B1 patent drawingFigure 3~4

AI summary

A messaging model and node induction methods and corresponding devices and systems are disclosed herein that are effective to enable as inductor node to induct: an inductee node into a distributed computing system and to enable the inducted node to carry out predetermined tasks.