Packet Network Performance Measurement with Identifiable Sample Packets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for performance measurement in packet-switched communication networks, such as those using the spin bit technique, fail to allow intermediate measurement points to identify specific sample packets for individual time measurements, particularly RTT measurements, and do not provide reliable correlation between counter-propagating flows.
Innovation Solution
A method where nodes in a packet-switched network alternately switch marking values in packets and write a sampling value in a single packet during each marking period, ensuring that a single couple of counter-propagating sample packets is exchanged, allowing measurement points to identify and measure one-way time and RTT accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If alternate marking technique is used to enable performance measurement, then packet loss and delay measurements can be performed, but intermediate measurement points cannot identify specific sample packets for individual time measurements
Solution Approach 1:
The patent segments the packet flow by introducing a sampling value field that identifies specific sample packets within the flow. This segmentation allows intermediate measurement points to distinguish and measure individual sample packets while maintaining the overall alternate marking structure for performance measurement.
Solution Approach 2:
The sampling value acts as an intermediary element that bridges the gap between the alternate marking technique and individual packet identification. By adding this intermediary field, measurement points can correlate sample packets across counter-propagating flows without requiring complex synchronization mechanisms.
2Adaptability or versatility
If spin bit technique is used for RTT measurements, then clock synchronization is not required, but reliable correlation between counter-propagating flows is not provided
Solution Approach 1:
The patent uses the sampling value as a copying mechanism where the same value is placed in packets of counter-propagating flows. This copying approach allows reliable correlation between flows without requiring clock synchronization, as the sampling value serves as a matching key that can be independently verified at measurement points.
3Quantity of substance
If multiple packets are marked as samples in each marking period, then more measurement data is available, but reception sequence errors cause measurement inaccuracies
Solution Approach 1:
The patent applies preliminary action by pre-calculating and pre-positioning sampling values in packets before transmission. This allows measurement points to identify sample packets based on predetermined sampling values, enabling accurate time measurements even when packets arrive out of sequence, as the sampling value provides a reliable identification marker independent of arrival order.
Data Source
Figure 1~2
Figure 3~4a
Figure 4b
AI summary
It is disclosed a method for enabling a performance measurement in a packet-switched communication network. A first node and a second node exchange packets comprising a marking value, which they alternately switch between two alternative marking values. The second node in particular switches the marking value applicable to the packets addressed to the other node depending on the marking value comprised in packets received therefrom. The first node writes a sampling value in one packet addressed to the second node for each marking period. Upon reception of each packet comprising the sampling value, the second node copies the sampling value in a packet addressed to the first node. One or more measurement points may be provided between the two nodes, which provide performance parameters for the packets comprising the sampling value in both directions. Such performance parameters may be used for providing round trip time measurements.