Packet Loss Measurement Using Trigger Packet Latching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for measuring packet loss in networks are limited by the need for continuous test packet generation, quiet periods, and alteration of existing network traffic, which restricts real-time measurements and requires synchronization of packet counts across probes.
Innovation Solution
A method and system that identify and latch packet counts at specific trigger packets across network locations, allowing for passive or active measurements without altering existing network traffic, using probes to monitor and filter packets based on common attributes and trigger filters to synchronize counts for accurate packet loss calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If test packets are generated continuously, then measurement coverage is improved, but packet count synchronization between probes deteriorates
Solution Approach 1:
A trigger packet is introduced as an intermediary mechanism to synchronize packet count measurements across multiple probes. The trigger packet acts as a reference point that all probes use to latch their respective packet counts, ensuring that counts correspond to the same packets even when probes are located at different positions in the network path.
2Object-affected harmful factors
If packet counts are requested out-of-band, then measurement interference is reduced, but measurement synchronization deteriorates
Solution Approach 1:
The measurement mechanism is merged with the data traffic itself by using in-band trigger packets that travel along the same network path as the packets being measured. This eliminates the need for separate out-of-band request channels, thereby maintaining measurement synchronization while minimizing additional network interference.
3Measurement precision
If test packets are generated in bursts with quiet periods, then packet count synchronization is improved, but real-time measurement capability deteriorates
Solution Approach 1:
The system enables continuous generation of test packets without requiring quiet periods between measurement cycles. By using trigger packets embedded within the continuous data flow, the system maintains both packet count synchronization and real-time measurement capability, eliminating the need to interrupt traffic for measurements.
4Measurement precision
If existing network traffic is altered by inserting packets or marking, then measurement accuracy is improved, but network traffic disruption deteriorates
Solution Approach 1:
The measurement system uses existing packets in the network traffic flow as trigger packets, allowing the traffic itself to serve the dual purpose of data transmission and measurement synchronization. This self-service approach eliminates the need to insert additional packets or modify existing ones, thereby maintaining measurement accuracy without disrupting network traffic.
Data Source
AI summary
In a method of measuring packet loss, a flow of packets is identified at first and second locations in a network. The packets in the flow are counted at each of the first and second locations to provide first and second packet counts, respectively. When a trigger packet in the flow is identified at each of the first and second locations, the first and second packet counts are latched to provide latched first and second packet counts corresponding to same packets in the flow. The latched first and second packet counts are compared to measure packet loss between the first and second locations.