Burst Estimator Logic for Network Event Visibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current telemetry systems face challenges in efficiently monitoring burstiness in network events due to bandwidth constraints and data storage limitations, leading to discarding and filtering of event data, which results in poor visibility and debuggability of network events.
Innovation Solution
A network device with burst estimator logic that tracks event rates over time periods, applying more weight to recent timestamps and less weight to older ones, allowing for the calculation of burstiness without storing all event timestamps, thereby reducing memory usage and enabling efficient burstiness measurement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all event timestamps are stored to calculate burstiness, then measurement precision of burstiness is improved, but device complexity and memory usage increase
Solution Approach 1:
The patent extracts only the essential information needed for burstiness calculation (rate of occurrence and time differences) from the complete event data (timestamps). Instead of storing all timestamps, the system extracts and stores only the rate of occurrence of events and the time difference between consecutive events, which is sufficient to calculate burstiness without retaining the full timestamp history.
Solution Approach 2:
The patent applies local quality by treating different aspects of event data differently. Rather than uniformly storing all timestamps with equal importance, the system focuses on local characteristics - the rate of occurrence and time differences between consecutive events - which are the critical local qualities needed for burstiness measurement. This selective focus reduces storage requirements while maintaining measurement accuracy.
2Loss of information
If event data is filtered to reduce data volume, then loss of information is reduced, but measurement precision of burstiness deteriorates
Solution Approach 1:
The patent extracts the critical features needed for burstiness calculation from the event data stream. Instead of filtering out data or storing all data, the system extracts the rate of occurrence and time differences between events, which are the essential features for burstiness measurement. This extraction approach minimizes information loss while avoiding the need to store complete event data.
Solution Approach 2:
The patent transforms the raw event timestamps into different parameters - specifically the rate of occurrence and time differences between consecutive events. This parameter transformation allows the system to work with derived values that capture the essential burstiness characteristics without requiring the original timestamp data, thereby reducing storage needs while preserving measurement precision.
Data Source
AI summary
In one embodiment, a network device, includes at least one network interface port to serve as at least one ingress port for receiving network packets from a network and as at least one egress port for forwarding network packets to the network, a memory, control circuitry configured to monitor respective events in the network device, the respective events having respective event timestamps, and burst estimator logic to track respective rates of occurrence of ones of the events over respective time periods responsively to respective ones of the event timestamps, follow a maximum one of the rates of occurrence, store, in the memory, the maximum rate of occurrence, and find a measure of burstiness of the events responsively to the followed maximum rate of occurrence.