Multicast Tracing With Hop-Level Latency Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multicast tracing tools do not support latency detection at individual hops of a multicast path, making it difficult to identify network devices causing delays, which can lead to degraded performance during channel switching in video streaming services like IPTV.
Innovation Solution
Enhance multicast tracing (mtrace) to incorporate latency information by determining local, network, and round-trip latencies at each network device using a flow sampling mechanism like IPFIX, allowing administrators to identify and address congestion points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing multicast tracing tools are used, then basic path tracing is available, but latency detection at individual hops is not supported
Solution Approach 1:
The patent segments the multicast path into individual hops (network devices) and measures latency at each hop separately. By dividing the overall path into discrete segments and tracking timestamps at each segment boundary, the system achieves precise latency detection at individual hops, resolving the contradiction between basic tracing and hop-specific measurement.
Solution Approach 2:
The patent introduces timestamping mechanisms and flow sampling (IPFIX) as intermediary elements between network devices and the tracing system. These intermediaries capture and transmit latency information without disrupting the multicast flow, enabling precise measurement while maintaining system functionality.
2Reliability
If latency information is collected at each network device, then congestion identification improves, but system complexity increases
Solution Approach 1:
Network devices autonomously generate and attach their own latency information (timestamps) to packets as they process them. Each device performs self-measurement without requiring external intervention or complex coordination, reducing overall system complexity while improving reliability through distributed self-monitoring.
Solution Approach 2:
The patent changes the measurement parameter from binary packet forwarding to timestamp-based latency calculation. By recording arrival and departure timestamps and computing the difference, the system transforms simple routing information into precise latency data without adding complex hardware, maintaining device simplicity while improving measurement capability.
3Measurement precision
If flow sampling mechanism like IPFIX is used, then latency measurement becomes possible, but implementation complexity increases
Solution Approach 1:
The patent leverages IPFIX (Internet Protocol Flow Information Export), an existing standard already deployed for traffic accounting and billing purposes. By making IPFIX serve dual purposes (billing + latency measurement), the system avoids implementing separate measurement infrastructure, reducing implementation complexity while enabling precise latency measurement.
Data Source
AI summary
A multicast tracing system for a network is provided. During operation, the system can identify, at a network device in the network, a join request and a data packet of a multicast group received at the network device. The system can determine respective timestamps indicating the arrival and departure of the join request and the data packet. The system can then determine, based on the timestamps, at least a local latency incurred by the data packet at the network device and a network latency incurred by the data packet in the network. Subsequently, the system can receive a multicast trace (mtrace) request for the multicast group. The system can incorporate the local latency and the network latency into the mtrace request. Upon identifying an upstream network device in a multicast path associated with the multicast group, the system can forward the mtrace request to the upstream network device.