Network Monitoring via Client-Network Signal Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large computer networks, such as those used in cloud computing, it is challenging to determine the expected network latency and identify the source of packet loss issues due to dynamic network topology changes and overlapping responsibilities between cloud providers, private network segments, and client-side components.
Innovation Solution
A monitoring system that transmits probes between user-defined endpoints to measure packet loss and round-trip time, comparing these metrics against network-based monitoring data to determine whether issues are within the cloud provider's network or client-side components, and provides signals to clients indicating the source of performance degradation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If network monitoring uses only network-based signals, then the monitoring coverage is comprehensive, but the ability to identify client-side issues is insufficient
Solution Approach 1:
The monitoring system is segmented into two independent components: network-based monitoring (performed by the cloud provider) and client-based monitoring (performed by the client device). Each component independently collects metrics from its respective perspective, allowing the system to maintain comprehensive coverage while improving issue identification accuracy through later correlation of the segmented data sets.
Solution Approach 2:
The patent introduces an intermediary correlation process that compares network-based metrics with client-based metrics. This intermediary layer acts as a mediator to determine whether performance issues originate from the cloud provider's network or the client-side components, resolving the contradiction between comprehensive monitoring and accurate issue attribution.
2Adaptability or versatility
If network topology changes dynamically, then the network adaptability is improved, but the difficulty of determining expected latency increases
Solution Approach 1:
The system performs preliminary action by continuously collecting and storing baseline network metrics and client-experienced metrics over time. This preliminary data collection establishes a historical record that enables comparison against current performance, allowing the system to determine expected latency ranges even as network topology changes dynamically.
Solution Approach 2:
The patent implements feedback by continuously comparing current network metrics with historical baseline data and client-reported metrics. This feedback mechanism allows the system to dynamically adjust expectations for latency and performance based on observed patterns, maintaining accurate performance determination despite network topology changes.
3Adaptability or versatility
If multiple sources contribute to performance issues, then the network functionality is enhanced, but the difficulty of identifying the specific source increases
Solution Approach 1:
The patent applies local quality by having different monitoring components operate with different measurement perspectives and characteristics. Network-based monitoring captures infrastructure-level metrics with one set of qualities, while client-based monitoring captures application-level metrics with different qualities. By maintaining these distinct local measurement qualities and comparing them, the system can precisely identify which specific source (cloud provider or client) is responsible for performance issues.
Data Source
AI summary
A monitoring service analyzes client-based monitoring data in correlation with network-based monitoring data for paths between two endpoints in a network upon request from a client. The endpoint can be any valid private IP address or DNS name where traffic is to be sent. Clients may define parameters for the monitoring, including thresholds for identifying network issues based on the client-based monitoring data. Different levels of network-based monitoring may be performed responsive to results of the correlation and continued performance determinations relative to the thresholds. The client-based and network-based monitoring data may include reports of performance signals, such as packet loss and round-trip time or other measured latency, as well as events such as connection issues and timeouts. The monitoring data from the different sources can be compared to determine whether a network issue is present in the network or outside of the network/in the client.