End-to-End Service Level Metrics from User-Side Anomaly Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud service providers face challenges in measuring end-to-end performance metrics for cloud applications due to limitations in monitoring beyond their infrastructure, leading to potential misattribution of user experience issues.
Innovation Solution
Implementing a service level component that collects end-to-end measurements from user devices, detects anomalies, and aggregates data to determine a service level metric, enabling identification of degraded applications and resource outages within a wide area network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If service level monitoring is performed only at cloud provider resources, then monitoring complexity is reduced, but measurement precision deteriorates because end-to-end user experience cannot be accurately assessed
Solution Approach 1:
The patent introduces an intermediary component that collects measurements from multiple sources including cloud provider resources and external network elements. This intermediary aggregates data from diverse points in the service delivery chain, enabling comprehensive end-to-end monitoring without requiring direct integration with every component, thus maintaining measurement precision while managing system complexity
Solution Approach 2:
The monitoring system is designed to perform multiple functions: collecting measurements from cloud resources, gathering user experience data from external sources, detecting anomalies, and generating service level reports. This multi-functional approach consolidates what would otherwise require separate specialized systems, improving measurement accuracy without proportionally increasing complexity
2Measurement precision
If end-to-end measurements are collected from user devices, then service level measurement accuracy is improved, but loss of information increases due to the complexity of multi-service communication paths
Solution Approach 1:
The patent segments the service delivery path into distinct measurable components, each monitored at appropriate points. By breaking down the end-to-end journey into discrete segments (cloud resources, network intermediaries, user device), the system can measure performance at each segment and aggregate the results, maintaining measurement precision while enabling systematic analysis of performance degradation sources
Solution Approach 2:
The monitoring system implements feedback mechanisms that continuously collect measurement data and use it to detect anomalies and attribute performance issues. The system feeds measurement results back through the analysis pipeline, enabling automated detection of degradation sources and reducing information loss about performance attribution by systematically processing and interpreting the collected data
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Described are examples for providing service level monitoring for a network hosting applications as a cloud service. A service level monitoring device may receive end-to-end measurements of service usage collected at user devices for a plurality of applications hosted as a cloud services. The service level monitoring device may determine degraded applications of the plurality of applications based on anomalies in the measurements. The service level monitoring device may determine a service level metric based on an aggregation of the degraded applications. In some examples, the service level monitoring device may detect a network outage affecting the service.