Network Volatility Monitoring Using Peer Device Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network monitoring tools struggle to accurately evaluate and predict network volatility, particularly short bursts and spikes that do not exceed operational thresholds, leading to unpredictable performance and user experience.
Innovation Solution
A system that measures volatility using coefficients of variation across multiple devices, identifies peer devices based on shared characteristics, and provides a dashboard for ranking and displaying network conditions, allowing for proactive identification and remediation of issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current monitoring tools use time-over-threshold windows and baseline deviation to identify abnormal performance, then they can detect sustained issues, but they fail to detect short spikes and bursts that do not exceed thresholds
Solution Approach 1:
The patent changes the monitoring parameter from absolute threshold-based detection to volatility-based detection using coefficient of variation. This allows the system to detect short spikes and bursts by measuring relative variability rather than absolute values, making previously invisible volatility visible while maintaining sensitivity to sustained issues.
Solution Approach 2:
The patent adds a new dimension of analysis by introducing volatility measurement as a separate metric from traditional threshold-based monitoring. This dimensional shift enables detection of short-term variations that exist in a different measurement space than traditional monitoring, allowing simultaneous detection of both sustained issues and short bursts.
2Reliability
If monitoring tools focus on metrics exceeding operational thresholds, then they can identify clear failures, but they cannot evaluate and predict network volatility accurately
Solution Approach 1:
The patent performs preliminary action by calculating and storing volatility metrics continuously, even before abnormal conditions occur. This pre-computation of volatility data enables accurate prediction and evaluation of network behavior patterns, allowing the system to anticipate potential issues before they manifest as threshold violations.
Solution Approach 2:
The patent implements feedback by using volatility measurements to continuously refine the understanding of normal versus abnormal network behavior. The system feeds volatility data back into the monitoring model, improving its ability to predict future performance and distinguish between acceptable variations and genuine anomalies.
3Loss of information
If monitoring tools use normalized deviations to widen the normal baseline window, then they can reduce false positives, but they eliminate events and prevent future deviations from being recognized
Solution Approach 1:
The patent segments the monitoring approach into two independent components: volatility measurement for detecting variations and threshold-based monitoring for identifying critical events. This segmentation allows the system to reduce false positives through volatility analysis while maintaining separate, sensitive event detection capabilities that are not suppressed by baseline adjustments.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
Novel solutions for monitoring and analyzing networks in terms of the volatility of various devices. Some solutions consider a weighted set of metrics in determining such volatility. Evaluation of devices against peers in view of these factors can produce insight about network conditions.