Modified Consensus Protocol Eliminates Heartbeat Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems face challenges in efficiently reestablishing communication after network or node failures, particularly in large-scale systems where continuous connections cannot be guaranteed, leading to issues with detection of communication failures and ensuring data consistency across nodes.
Innovation Solution
A method involving a central topology manager to detect node liveness and reinitiate communication, using communication resumption information such as invocation counts and last message indices to determine whether communication can be safely resumed without losing messages, and implementing a modified consensus protocol to eliminate heartbeat network traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional consensus protocols with heartbeat messages are used to detect node liveness, then node failure detection capability is improved, but network traffic increases significantly
Solution Approach 1:
The patent extracts the heartbeat mechanism from the consensus protocol entirely. Instead of nodes exchanging periodic heartbeat messages to detect liveness, the system removes this mechanism and relies on the natural absence of communication to detect failures. This eliminates the unnecessary network traffic while maintaining failure detection capability through the central topology manager that tracks node communication patterns.
2Measurement precision
If nodes continuously monitor communication to detect failures in large-scale distributed systems, then failure detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces a central topology manager as an intermediary that consolidates the failure detection function. Instead of each node independently monitoring and making decisions about other nodes, the central topology manager receives communication information from all nodes and determines liveness centrally. This reduces system complexity by centralizing the monitoring logic while maintaining accurate failure detection through aggregated communication data.
3Productivity
If communication is reestablished after detected failures without verification, then communication recovery speed is improved, but data consistency is compromised
Solution Approach 1:
The patent applies preliminary action by having nodes verify liveness status through the central topology manager before attempting to reestablish communication. The system checks whether a node is truly alive or merely experiencing temporary network issues before resuming communication. This preliminary verification step ensures data consistency is maintained while enabling relatively quick recovery once liveness is confirmed, preventing premature reconnection that could cause inconsistency.
Data Source
AI summary
A computing system in data communication with a plurality of nodes that make up a distributed computing cluster can detect an absence of communication from a node of the plurality of nodes over a time period that exceeds a predefined threshold time period. The computing system can query an instance of a central topology manager for the plurality of nodes regarding liveness of the node from which the absence of communication was detected and can attempting to re-initiate communication with the node when the instance of the central topology manager indicates that the node is live.