Lightweight Multicast Congestion Detection via Negative Acknowledgment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data distribution services in large-scale networks face reliability issues due to congestion, which can be exacerbated by simple packet retransmission methods, leading to performance degradation and potential congestion problems.
Innovation Solution
A lightweight multicast method and apparatus that detects network congestion using negative response messages from subscriber nodes, determines the cause of congestion through cross-layered congestion tracking, and employs recovery control policies such as naive, fast, and proactive recovery to alleviate congestion, minimizing control messages and improving reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If packet retransmission is used to ensure reliability, then delivery reliability is improved, but system performance deteriorates and congestion increases
Solution Approach 1:
The patent implements a feedback mechanism where subscriber nodes send negative acknowledgment (NAK) messages to publisher nodes when packet loss is detected. This feedback loop enables the system to identify congestion issues and trigger appropriate recovery actions, resolving the contradiction between ensuring reliable delivery and maintaining system performance by only retransmitting when necessary.
Solution Approach 2:
The patent changes the parameter of packet transmission from simple retransmission to selective retransmission based on congestion detection. By monitoring network conditions and only retransmitting packets when congestion is detected, the system maintains reliability while avoiding the performance degradation associated with continuous retransmission.
2Reliability
If simple packet retransmission is used, then reliability is improved, but network congestion is exacerbated
Solution Approach 1:
The patent uses feedback from subscriber nodes to publisher nodes through NAK messages to detect congestion conditions. This feedback mechanism enables the system to distinguish between successful transmissions and congestion-induced losses, allowing selective retransmission that prevents exacerbating network congestion while maintaining delivery reliability.
Solution Approach 2:
The patent introduces an intermediary congestion detection mechanism that acts as a mediator between packet transmission and retransmission decisions. By detecting congestion through NAK messages and using this information to control retransmission behavior, the system prevents the harmful effect of network congestion while ensuring reliable data delivery.
3Measurement precision
If additional messages are used for congestion detection, then congestion detection accuracy is improved, but control message overhead increases
Solution Approach 1:
The patent implements self-service by having subscriber nodes generate and send NAK messages automatically when packet loss is detected, without requiring additional dedicated congestion detection messages. This approach provides accurate congestion detection while minimizing control message overhead, as the same NAK messages serve both error notification and congestion detection purposes.
Solution Approach 2:
The patent makes the NAK message universal by using it for multiple purposes: confirming packet loss detection and indicating congestion conditions. This multi-functional approach enables accurate congestion detection without increasing control message overhead, as the same message type serves both error feedback and congestion notification functions.
Data Source
AI summary
The present invention provides a lightweight multicast method and apparatus for a data distribution service. The lightweight multicast apparatus includes a network congestion detection unit for multicasting packets to a plurality of subscriber node terminals, and detecting a cause of network congestion using a negative response message transferred from a relevant subscriber node terminal that could not receive the packets among the plurality of subscriber node terminals. A network recovery control unit determines a recovery control policy depending on the cause of the congestion detected by the network congestion detection unit, and solves the cause of the congestion.