Arbitrator Delay Mechanism for Distributed Node Continuity

Solution Overview

Problem

Existing failure detection schemes in distributed computing systems can cause nodes to terminate prematurely or unnecessarily due to network congestion or failure, leading to a reduction in processing and storage resources.

Innovation Solution

Implementing delay and neutral arbitration results in lease protocols to allow both nodes to continue operating by utilizing a waiting period for arbitration requests, preventing unnecessary termination and allowing time for requests to reach the arbitrator.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional arbitration schemes immediately grant or reject arbitration requests upon receipt, then the arbitration process is fast and decisive, but nodes may terminate prematurely or unnecessarily due to network congestion or temporary communication failures

Engineering Contradiction:

Improvenode operation continuityVSAvoidarbitration decision delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The arbitrator performs preliminary actions by issuing delay arbitration results and starting waiting periods before making final decisions. This allows the system to prepare for potential network issues by giving nodes temporary continuation permission and monitoring for additional arbitration requests during the waiting period, thereby preventing premature termination while maintaining timely response

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system cushions against network congestion and temporary failures by implementing a waiting period mechanism that allows nodes to continue operating during arbitration uncertainty. This cushioning approach prevents unnecessary node terminations caused by transient network issues while still enabling decisive action when failures are confirmed

Inventive Principle:

Principle #11Beforehand cushioning (Prior cushioning)

2Reliability

If the arbitrator waits for multiple arbitration requests before making decisions, then unnecessary node terminations are prevented, but the arbitration process becomes more complex and requires tracking multiple requests and timing

Engineering Contradiction:

Improvenode operation continuityVSAvoidarbitration process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The arbitrator performs preliminary actions by issuing delay arbitration results and starting waiting periods before making final decisions. This allows the system to prepare for potential network issues by giving nodes temporary continuation permission and monitoring for additional arbitration requests during the waiting period, thereby preventing premature termination while maintaining timely response

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The arbitration process is made dynamic by implementing time-based state transitions. The system automatically transitions from an initial arbitration request state to a waiting period state, and then to a final decision state based on elapsed time and receipt of additional requests. This dynamic approach manages complexity through automated temporal logic rather than complex manual coordination

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS10055315B2Failure monitoring in distributed computing systems

Publication Date: 2018.08.21 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Arbitration techniques in distributed computing systems are disclosed herein. In one embodiment, a method includes receiving an arbitration request from a first node in the computing system. The first arbitration request indicating that a one-way lease between the first node and a second node in the computing system has failed. The method also includes in response to receiving the arbitration request from the first node, providing an arbitration result to the first node, the arbitration result allowing both the first and second nodes to continue to operate despite of that the one-way lease between the first and second nodes has failed.