Client Anomaly Isolation Using Normalized Characteristic Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software vendors face challenges in debugging issues within customer networks due to limited visibility, leading to inaccurate information and potential privacy concerns, which can result in a negative customer perception.
Innovation Solution
A method and apparatus for detecting and isolating client application anomalies by comparing client characteristics of abnormal and normal clients, using pre-processing to define norms, adjusting for seasonality, and normalizing metrics to identify and isolate client-side anomalies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If remote access is requested to debug customer network issues, then visibility into the problem is improved, but customer privacy concerns increase and customer acceptance decreases
Solution Approach 1:
The patent creates a virtual copy of the customer's network environment by deploying an agent on their system. This agent captures and transmits relevant data, logs, and diagnostic information to the vendor's analysis system, enabling remote debugging without requiring direct vendor access to the customer's physical network infrastructure.
Solution Approach 2:
The patent introduces an intermediary agent software component that sits between the customer's network and the vendor's debugging system. This agent acts as a mediator that collects, processes, and transmits only necessary diagnostic information while maintaining customer control over what data is shared, thus addressing privacy concerns.
2Difficulty of detecting and measuring
If remote access is requested for debugging, then issue diagnosis capability is improved, but customer acceptance and trust deteriorate
Solution Approach 1:
The patent enables the customer's system to perform self-diagnosis by running local agent software that automatically collects diagnostic data, identifies issues, and transmits findings to the vendor. This self-service approach empowers the customer's system to participate actively in the debugging process without requiring vendor control.
Solution Approach 2:
The patent segments the debugging functionality into separate components: a lightweight agent on the customer's system that handles data collection, and a separate analysis system at the vendor's end that processes the information. This segmentation allows the customer to maintain control of their system while still enabling comprehensive diagnosis.
3Measurement precision
If comprehensive customer network data is collected for analysis, then debugging accuracy is improved, but data processing complexity and time increase
Solution Approach 1:
The patent applies local quality by having the customer's system collect and pre-process only the specific diagnostic data relevant to the reported issue, rather than gathering all possible network data. The agent intelligently filters and prioritizes information based on the nature of the problem being investigated.
Solution Approach 2:
The patent performs preliminary data processing and filtering at the customer's system before transmission to the vendor. The agent pre-processes logs, aggregates relevant metrics, and prepares structured diagnostic information in advance, reducing the complexity of subsequent analysis at the vendor's end.
Data Source
AI summary
A method for detecting and isolating client application anomalies includes detecting an abnormal situation occurring in a software and determining, a set of abnormal clients and a set of normal clients in response to detecting the abnormal situation occurring in the software. The method includes comparing one or more client characteristics of the set of abnormal clients with one or more client characteristics of the set of normal clients and determining one or more client-side anomalies based, at least in part, on the comparison.