End-to-End Loss Distribution Estimation via Hop Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network monitoring methods, either active or passive, fail to accurately measure end-to-end packet loss, leading to incomplete or inaccurate views of network performance, which can impact service level agreements (SLAs) and user experiences.
Innovation Solution
A system that combines active and passive monitoring by measuring loss rates at each network hop and using statistical models to estimate end-to-end packet loss, incorporating a central network monitor that utilizes a path determination module, interface-based path loss determination module, and compliance determination module to provide a comprehensive view of network performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If active monitoring uses probe packets sent at low rate, then network disruption is minimized, but measurement precision of end-to-end loss deteriorates due to under-sampling
Solution Approach 1:
The patent segments the end-to-end path into multiple hops between adjacent network devices. Instead of measuring loss directly end-to-end with low-rate probes, the system measures loss at each individual hop segment and then combines these segment losses to estimate the total end-to-end loss. This segmentation allows for higher measurement resolution at each segment while maintaining network stability.
Solution Approach 2:
The patent performs measurements at each network device hop, which is more granular than traditional end-to-end probing. By measuring loss partially at each intermediate node rather than only at the destination, the system achieves better measurement precision without requiring excessive probe traffic, thus maintaining network stability.
2Measurement precision
If passive monitoring collects statistics at each network device, then loss rate measurement at individual devices is accurate, but end-to-end path loss information is lost
Solution Approach 1:
The patent uses a feedback mechanism where each network device reports its measured loss statistics back to a central controller. The controller then combines these feedback reports from multiple devices to reconstruct the end-to-end loss distribution. This feedback loop preserves both the detailed device-level measurements and the aggregate end-to-end path information.
Solution Approach 2:
The patent merges the loss measurements from multiple network devices along a path by combining their individual loss distributions using convolution operations. This merging process integrates the accurate device-level statistics into a comprehensive end-to-end loss estimate, preserving information from all segments.
3Measurement precision
If probe packets are sent at high rate for accurate measurement, then measurement precision improves, but network disruption increases
Solution Approach 1:
By segmenting measurements into small hop-by-hop segments, the system achieves accurate measurement without requiring high rates of end-to-end probe packets. Each segment uses local measurements that can be performed at lower rates, reducing overall network disruption while maintaining precision through the combination of segments.
Solution Approach 2:
The patent enables each network device to perform self-measurement of packet loss at its own interfaces using locally available statistics. This self-service approach eliminates the need for external high-rate probe traffic, as each device independently measures its own loss characteristics and reports them to the controller.
Data Source
AI summary
In computing networks, end-to-end loss distributions are estimated according to the present disclosure. The loss distributions may be calculated based on loss values reported by nodes in the network along an end-to-end communication path, which may be identified by a route tracing process. The loss values for each node in the path for a given time window may be reduced by a rounding process, and the reduced values for adjacent nodes in the path may be iteratively multiplied with one another in a pairwise manner to generate an end-to-end loss value vector, where the resulting product from each iteration may be reduced by the rounding process. A graphical or other representation of the values in the end-to-end loss value vector indicating a distribution of losses experienced along the path may be output via a user interface.